From dddc5f1c39fa9750f1df1635be6e208697e85357 Mon Sep 17 00:00:00 2001
From: Dimitri Lozeve <dimitri@lozeve.com>
Date: Tue, 5 May 2020 09:55:33 +0200
Subject: [PATCH 1/6] Add ICLR 2020 Notes

---
 _site/archive.html               |   4 +
 _site/atom.xml                   | 111 ++++++++++++++++++++-
 _site/index.html                 |   4 +
 _site/posts/iclr-2020-notes.html | 155 +++++++++++++++++++++++++++++
 _site/rss.xml                    | 111 ++++++++++++++++++++-
 posts/iclr-2020-notes.org        | 162 +++++++++++++++++++++++++++++++
 6 files changed, 545 insertions(+), 2 deletions(-)
 create mode 100644 _site/posts/iclr-2020-notes.html
 create mode 100644 posts/iclr-2020-notes.org

diff --git a/_site/archive.html b/_site/archive.html
index cdb6910..fefe245 100644
--- a/_site/archive.html
+++ b/_site/archive.html
@@ -47,6 +47,10 @@
     Here you can find all my previous posts:
 <ul>
     
+        <li>
+            <a href="./posts/iclr-2020-notes.html">ICLR 2020 Notes</a> - May  5, 2020
+        </li>
+    
         <li>
             <a href="./posts/hierarchical-optimal-transport-for-document-classification.html">Reading notes: Hierarchical Optimal Transport for Document Representation</a> - April  5, 2020
         </li>
diff --git a/_site/atom.xml b/_site/atom.xml
index c20090d..bd3c5d0 100644
--- a/_site/atom.xml
+++ b/_site/atom.xml
@@ -8,8 +8,117 @@
         <name>Dimitri Lozeve</name>
         <email>dimitri+web@lozeve.com</email>
     </author>
-    <updated>2020-04-05T00:00:00Z</updated>
+    <updated>2020-05-05T00:00:00Z</updated>
     <entry>
+    <title>ICLR 2020 Notes</title>
+    <link href="https://www.lozeve.com/posts/iclr-2020-notes.html" />
+    <id>https://www.lozeve.com/posts/iclr-2020-notes.html</id>
+    <published>2020-05-05T00:00:00Z</published>
+    <updated>2020-05-05T00:00:00Z</updated>
+    <summary type="html"><![CDATA[<article>
+    <section class="header">
+        
+    </section>
+    <section>
+        <p>ICLR is one of the most important conferences in machine learning, and as such, I was very excited to have the opportunity to volunteer and attend the first fully-virtual edition of the event. The whole content of the conference has been made <a href="https://iclr.cc/virtual_2020/index.html">publicly available</a>, only a few days after the end of the event!</p>
+<p>I would like to thank the <a href="https://iclr.cc/Conferences/2020/Committees">organizing committee</a> for this incredible event, and the possibility to volunteer to help other participants<span><label for="sn-1" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-1" class="margin-toggle"/><span class="sidenote">To better organize the event, and help people navigate the various online tools, they brought in 500(!) volunteers, waved our registration fees, and asked us to do simple load-testing and tech support. This was a very generous offer, and felt very rewarding for us, as we could attend the conference, and give back to the organization a little bit.<br />
+<br />
+</span></span>.</p>
+<p>The many volunteers, the online-only nature of the event, and the low registration fees also allowed for what felt like a very diverse, inclusive event. Many graduate students and researchers from industry (like me), who do not generally have the time or the resources to travel to conferences like this, were able to attend, and make the exchanges richer.</p>
+<p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
+<h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
+<p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
+<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
+<br />
+</span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
+<p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
+<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
+<h1 id="speakers">Speakers</h1>
+<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
+<h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
+<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter.</p>
+<h1 id="workshops">Workshops</h1>
+<h1 id="some-interesting-papers">Some Interesting Papers</h1>
+<h2 id="natural-language-processing">Natural Language Processing</h2>
+<h3 id="harwath-et-al.-learning-hierarchical-discrete-linguistic-units-from-visually-grounded-speech">Harwath et al., <a href="https://openreview.net/forum?id=B1elCp4KwH">Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech</a></h3>
+<p>Humans can easily deconstruct all the information available in speech (meaning, language, emotion, speaker, etc.). However, this is very hard for machines. This paper explores the capacity of algorithms to reason about all the aspects of the signal, including visual cues.</p>
+<p>Their goal is to use spoken captions of images to train a predictive model.</p>
+<ul>
+<li>speech signal: contains a lot of information (meaning, language, emotion, speaker, environment)
+<ul>
+<li>difficult to identify these different parts for an algo (although easy for a human)</li>
+</ul></li>
+<li>dominated by supervised ML
+<ul>
+<li>automated speech recognition (ASR) = P(text | waveform)</li>
+<li>text-to-speech (TTS) = P(waveform | text, speaker)</li>
+<li>high sample complexity</li>
+<li>bad out-of-domain performance</li>
+<li>limited by annotation capability</li>
+</ul></li>
+<li>human-like learning
+<ul>
+<li>ability to jointly reason about all the aspects of the signal</li>
+<li>rapidly adapt to new speaker or noise conditions</li>
+<li>learn new words for a single example</li>
+<li>utilize unlabelled multimodal data</li>
+</ul></li>
+<li>using visual grounding for self-supervision
+<ul>
+<li>humans can leverage cross-modal correspondences to learn what spoken words represent without requiring any text or symbolic input whatsoever</li>
+<li>hypothesis: similar for computer algorithms?</li>
+</ul></li>
+<li>goal: use spoken captions of images to train a predictive model
+<ul>
+<li>learn a hierarchical structure of units</li>
+<li>learn the corresponding text, but also the transcription of the spoken sounds, at a sub-word level</li>
+</ul></li>
+<li>prefer models that learn a discrete tokenisation of the speech
+<ul>
+<li>language has an intrinsically symbolic structure
+<ul>
+<li>convey meaning with discrete words</li>
+<li>words are in turn composed of a finite set of speech sounds (phones)</li>
+</ul></li>
+<li>model that can discover discrete representations for word and phone-like units
+<ul>
+<li>more interpretable</li>
+<li>able to do few-shot learning (learn a new word-like unit in terms of known phone-like units)</li>
+<li>path towards learning compositional structure from continuous signals</li>
+</ul></li>
+</ul></li>
+<li>model for audio-visual grounding:
+<ul>
+<li>NN for image</li>
+<li>NN for raw speech</li>
+<li>shared embedding space</li>
+<li>semantic supervision</li>
+</ul></li>
+<li>preliminary studies
+<ul>
+<li>lower layers features are correlated with phones</li>
+<li>higher lauer features are correlated with words</li>
+</ul></li>
+<li>add a vector quantizing layer in the speech NN</li>
+<li>hierarchy of quantization layers
+<ul>
+<li>capture phones and words</li>
+</ul></li>
+<li>do an ABX test to compare performance to speech-only models</li>
+<li>conclusion
+<ul>
+<li>novel linguistic unit learning paradigm using multimodal data without text</li>
+<li>SOTA performance on learning phonetic and word-level units</li>
+<li>discovery of discreteness as a good inductive bias for semantic task from speech</li>
+</ul></li>
+</ul>
+<h2 id="reinforcement-learning">Reinforcement Learning</h2>
+<h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
+    </section>
+</article>
+]]></summary>
+</entry>
+<entry>
     <title>Reading notes: Hierarchical Optimal Transport for Document Representation</title>
     <link href="https://www.lozeve.com/posts/hierarchical-optimal-transport-for-document-classification.html" />
     <id>https://www.lozeve.com/posts/hierarchical-optimal-transport-for-document-classification.html</id>
diff --git a/_site/index.html b/_site/index.html
index f4ee194..87f633e 100644
--- a/_site/index.html
+++ b/_site/index.html
@@ -73,6 +73,10 @@ public key: RWQ6uexORp8f7USHA7nX9lFfltaCA9x6aBV06MvgiGjUt6BVf6McyD26
 
 <ul>
     
+        <li>
+            <a href="./posts/iclr-2020-notes.html">ICLR 2020 Notes</a> - May  5, 2020
+        </li>
+    
         <li>
             <a href="./posts/hierarchical-optimal-transport-for-document-classification.html">Reading notes: Hierarchical Optimal Transport for Document Representation</a> - April  5, 2020
         </li>
diff --git a/_site/posts/iclr-2020-notes.html b/_site/posts/iclr-2020-notes.html
new file mode 100644
index 0000000..3281e1e
--- /dev/null
+++ b/_site/posts/iclr-2020-notes.html
@@ -0,0 +1,155 @@
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8">
+    <meta http-equiv="x-ua-compatible" content="ie=edge">
+    <meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=yes">
+    <meta name="description" content="Dimitri Lozeve's blog: ICLR 2020 Notes">
+    
+    <title>Dimitri Lozeve - ICLR 2020 Notes</title>
+    <link rel="stylesheet" href="../css/tufte.css" />
+    <link rel="stylesheet" href="../css/pandoc.css" />
+    <link rel="stylesheet" href="../css/default.css" />
+    <link rel="stylesheet" href="../css/syntax.css" />
+
+    <!-- KaTeX CSS styles -->
+    <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/katex@0.11.0/dist/katex.min.css" integrity="sha384-BdGj8xC2eZkQaxoQ8nSLefg4AV4/AwB3Fj+8SUSo7pnKP6Eoy18liIKTPn9oBYNG" crossorigin="anonymous">
+
+    <!-- The loading of KaTeX is deferred to speed up page rendering -->
+    <script defer src="https://cdn.jsdelivr.net/npm/katex@0.11.0/dist/katex.min.js" integrity="sha384-JiKN5O8x9Hhs/UE5cT5AAJqieYlOZbGT3CHws/y97o3ty4R7/O5poG9F3JoiOYw1" crossorigin="anonymous"></script>
+
+    <!-- To automatically render math in text elements, include the auto-render extension: -->
+    <script defer src="https://cdn.jsdelivr.net/npm/katex@0.11.0/dist/contrib/auto-render.min.js" integrity="sha384-kWPLUVMOks5AQFrykwIup5lo0m3iMkkHrD0uJ4H5cjeGihAutqP0yW0J6dpFiVkI" crossorigin="anonymous" onload="renderMathInElement(document.body);"></script>
+ 
+  </head>
+  <body>
+    <article>
+      
+      <header>
+	<nav>
+          <a href="../">Home</a>
+	  <a href="../projects.html">Projects</a>
+          <a href="../archive.html">Archive</a>
+	  <a href="../contact.html">Contact</a>
+	</nav>
+
+	<h1 class="title">ICLR 2020 Notes</h1>
+	
+	
+	<p class="byline">May  5, 2020</p>
+	
+      </header>
+      
+
+      
+    </article>
+
+    <article>
+    <section class="header">
+        
+    </section>
+    <section>
+        <p>ICLR is one of the most important conferences in machine learning, and as such, I was very excited to have the opportunity to volunteer and attend the first fully-virtual edition of the event. The whole content of the conference has been made <a href="https://iclr.cc/virtual_2020/index.html">publicly available</a>, only a few days after the end of the event!</p>
+<p>I would like to thank the <a href="https://iclr.cc/Conferences/2020/Committees">organizing committee</a> for this incredible event, and the possibility to volunteer to help other participants<span><label for="sn-1" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-1" class="margin-toggle" /><span class="sidenote">To better organize the event, and help people navigate the various online tools, they brought in 500(!) volunteers, waved our registration fees, and asked us to do simple load-testing and tech support. This was a very generous offer, and felt very rewarding for us, as we could attend the conference, and give back to the organization a little bit.<br />
+<br />
+</span></span>.</p>
+<p>The many volunteers, the online-only nature of the event, and the low registration fees also allowed for what felt like a very diverse, inclusive event. Many graduate students and researchers from industry (like me), who do not generally have the time or the resources to travel to conferences like this, were able to attend, and make the exchanges richer.</p>
+<p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
+<h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
+<p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
+<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle" /><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
+<br />
+</span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
+<p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
+<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
+<h1 id="speakers">Speakers</h1>
+<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
+<h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
+<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter.</p>
+<h1 id="workshops">Workshops</h1>
+<h1 id="some-interesting-papers">Some Interesting Papers</h1>
+<h2 id="natural-language-processing">Natural Language Processing</h2>
+<h3 id="harwath-et-al.-learning-hierarchical-discrete-linguistic-units-from-visually-grounded-speech">Harwath et al., <a href="https://openreview.net/forum?id=B1elCp4KwH">Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech</a></h3>
+<p>Humans can easily deconstruct all the information available in speech (meaning, language, emotion, speaker, etc.). However, this is very hard for machines. This paper explores the capacity of algorithms to reason about all the aspects of the signal, including visual cues.</p>
+<p>Their goal is to use spoken captions of images to train a predictive model.</p>
+<ul>
+<li>speech signal: contains a lot of information (meaning, language, emotion, speaker, environment)
+<ul>
+<li>difficult to identify these different parts for an algo (although easy for a human)</li>
+</ul></li>
+<li>dominated by supervised ML
+<ul>
+<li>automated speech recognition (ASR) = P(text | waveform)</li>
+<li>text-to-speech (TTS) = P(waveform | text, speaker)</li>
+<li>high sample complexity</li>
+<li>bad out-of-domain performance</li>
+<li>limited by annotation capability</li>
+</ul></li>
+<li>human-like learning
+<ul>
+<li>ability to jointly reason about all the aspects of the signal</li>
+<li>rapidly adapt to new speaker or noise conditions</li>
+<li>learn new words for a single example</li>
+<li>utilize unlabelled multimodal data</li>
+</ul></li>
+<li>using visual grounding for self-supervision
+<ul>
+<li>humans can leverage cross-modal correspondences to learn what spoken words represent without requiring any text or symbolic input whatsoever</li>
+<li>hypothesis: similar for computer algorithms?</li>
+</ul></li>
+<li>goal: use spoken captions of images to train a predictive model
+<ul>
+<li>learn a hierarchical structure of units</li>
+<li>learn the corresponding text, but also the transcription of the spoken sounds, at a sub-word level</li>
+</ul></li>
+<li>prefer models that learn a discrete tokenisation of the speech
+<ul>
+<li>language has an intrinsically symbolic structure
+<ul>
+<li>convey meaning with discrete words</li>
+<li>words are in turn composed of a finite set of speech sounds (phones)</li>
+</ul></li>
+<li>model that can discover discrete representations for word and phone-like units
+<ul>
+<li>more interpretable</li>
+<li>able to do few-shot learning (learn a new word-like unit in terms of known phone-like units)</li>
+<li>path towards learning compositional structure from continuous signals</li>
+</ul></li>
+</ul></li>
+<li>model for audio-visual grounding:
+<ul>
+<li>NN for image</li>
+<li>NN for raw speech</li>
+<li>shared embedding space</li>
+<li>semantic supervision</li>
+</ul></li>
+<li>preliminary studies
+<ul>
+<li>lower layers features are correlated with phones</li>
+<li>higher lauer features are correlated with words</li>
+</ul></li>
+<li>add a vector quantizing layer in the speech NN</li>
+<li>hierarchy of quantization layers
+<ul>
+<li>capture phones and words</li>
+</ul></li>
+<li>do an ABX test to compare performance to speech-only models</li>
+<li>conclusion
+<ul>
+<li>novel linguistic unit learning paradigm using multimodal data without text</li>
+<li>SOTA performance on learning phonetic and word-level units</li>
+<li>discovery of discreteness as a good inductive bias for semantic task from speech</li>
+</ul></li>
+</ul>
+<h2 id="reinforcement-learning">Reinforcement Learning</h2>
+<h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
+    </section>
+</article>
+
+
+    <footer>
+      Site proudly generated by
+      <a href="http://jaspervdj.be/hakyll">Hakyll</a>
+    </footer>
+  </body>
+</html>
diff --git a/_site/rss.xml b/_site/rss.xml
index d59e504..9b0802f 100644
--- a/_site/rss.xml
+++ b/_site/rss.xml
@@ -7,8 +7,117 @@
         <description><![CDATA[Recent posts]]></description>
         <atom:link href="https://www.lozeve.com/rss.xml" rel="self"
                    type="application/rss+xml" />
-        <lastBuildDate>Sun, 05 Apr 2020 00:00:00 UT</lastBuildDate>
+        <lastBuildDate>Tue, 05 May 2020 00:00:00 UT</lastBuildDate>
         <item>
+    <title>ICLR 2020 Notes</title>
+    <link>https://www.lozeve.com/posts/iclr-2020-notes.html</link>
+    <description><![CDATA[<article>
+    <section class="header">
+        
+    </section>
+    <section>
+        <p>ICLR is one of the most important conferences in machine learning, and as such, I was very excited to have the opportunity to volunteer and attend the first fully-virtual edition of the event. The whole content of the conference has been made <a href="https://iclr.cc/virtual_2020/index.html">publicly available</a>, only a few days after the end of the event!</p>
+<p>I would like to thank the <a href="https://iclr.cc/Conferences/2020/Committees">organizing committee</a> for this incredible event, and the possibility to volunteer to help other participants<span><label for="sn-1" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-1" class="margin-toggle"/><span class="sidenote">To better organize the event, and help people navigate the various online tools, they brought in 500(!) volunteers, waved our registration fees, and asked us to do simple load-testing and tech support. This was a very generous offer, and felt very rewarding for us, as we could attend the conference, and give back to the organization a little bit.<br />
+<br />
+</span></span>.</p>
+<p>The many volunteers, the online-only nature of the event, and the low registration fees also allowed for what felt like a very diverse, inclusive event. Many graduate students and researchers from industry (like me), who do not generally have the time or the resources to travel to conferences like this, were able to attend, and make the exchanges richer.</p>
+<p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
+<h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
+<p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
+<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
+<br />
+</span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
+<p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
+<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
+<h1 id="speakers">Speakers</h1>
+<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
+<h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
+<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter.</p>
+<h1 id="workshops">Workshops</h1>
+<h1 id="some-interesting-papers">Some Interesting Papers</h1>
+<h2 id="natural-language-processing">Natural Language Processing</h2>
+<h3 id="harwath-et-al.-learning-hierarchical-discrete-linguistic-units-from-visually-grounded-speech">Harwath et al., <a href="https://openreview.net/forum?id=B1elCp4KwH">Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech</a></h3>
+<p>Humans can easily deconstruct all the information available in speech (meaning, language, emotion, speaker, etc.). However, this is very hard for machines. This paper explores the capacity of algorithms to reason about all the aspects of the signal, including visual cues.</p>
+<p>Their goal is to use spoken captions of images to train a predictive model.</p>
+<ul>
+<li>speech signal: contains a lot of information (meaning, language, emotion, speaker, environment)
+<ul>
+<li>difficult to identify these different parts for an algo (although easy for a human)</li>
+</ul></li>
+<li>dominated by supervised ML
+<ul>
+<li>automated speech recognition (ASR) = P(text | waveform)</li>
+<li>text-to-speech (TTS) = P(waveform | text, speaker)</li>
+<li>high sample complexity</li>
+<li>bad out-of-domain performance</li>
+<li>limited by annotation capability</li>
+</ul></li>
+<li>human-like learning
+<ul>
+<li>ability to jointly reason about all the aspects of the signal</li>
+<li>rapidly adapt to new speaker or noise conditions</li>
+<li>learn new words for a single example</li>
+<li>utilize unlabelled multimodal data</li>
+</ul></li>
+<li>using visual grounding for self-supervision
+<ul>
+<li>humans can leverage cross-modal correspondences to learn what spoken words represent without requiring any text or symbolic input whatsoever</li>
+<li>hypothesis: similar for computer algorithms?</li>
+</ul></li>
+<li>goal: use spoken captions of images to train a predictive model
+<ul>
+<li>learn a hierarchical structure of units</li>
+<li>learn the corresponding text, but also the transcription of the spoken sounds, at a sub-word level</li>
+</ul></li>
+<li>prefer models that learn a discrete tokenisation of the speech
+<ul>
+<li>language has an intrinsically symbolic structure
+<ul>
+<li>convey meaning with discrete words</li>
+<li>words are in turn composed of a finite set of speech sounds (phones)</li>
+</ul></li>
+<li>model that can discover discrete representations for word and phone-like units
+<ul>
+<li>more interpretable</li>
+<li>able to do few-shot learning (learn a new word-like unit in terms of known phone-like units)</li>
+<li>path towards learning compositional structure from continuous signals</li>
+</ul></li>
+</ul></li>
+<li>model for audio-visual grounding:
+<ul>
+<li>NN for image</li>
+<li>NN for raw speech</li>
+<li>shared embedding space</li>
+<li>semantic supervision</li>
+</ul></li>
+<li>preliminary studies
+<ul>
+<li>lower layers features are correlated with phones</li>
+<li>higher lauer features are correlated with words</li>
+</ul></li>
+<li>add a vector quantizing layer in the speech NN</li>
+<li>hierarchy of quantization layers
+<ul>
+<li>capture phones and words</li>
+</ul></li>
+<li>do an ABX test to compare performance to speech-only models</li>
+<li>conclusion
+<ul>
+<li>novel linguistic unit learning paradigm using multimodal data without text</li>
+<li>SOTA performance on learning phonetic and word-level units</li>
+<li>discovery of discreteness as a good inductive bias for semantic task from speech</li>
+</ul></li>
+</ul>
+<h2 id="reinforcement-learning">Reinforcement Learning</h2>
+<h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
+    </section>
+</article>
+]]></description>
+    <pubDate>Tue, 05 May 2020 00:00:00 UT</pubDate>
+    <guid>https://www.lozeve.com/posts/iclr-2020-notes.html</guid>
+    <dc:creator>Dimitri Lozeve</dc:creator>
+</item>
+<item>
     <title>Reading notes: Hierarchical Optimal Transport for Document Representation</title>
     <link>https://www.lozeve.com/posts/hierarchical-optimal-transport-for-document-classification.html</link>
     <description><![CDATA[<article>
diff --git a/posts/iclr-2020-notes.org b/posts/iclr-2020-notes.org
new file mode 100644
index 0000000..cd53eea
--- /dev/null
+++ b/posts/iclr-2020-notes.org
@@ -0,0 +1,162 @@
+---
+title: "ICLR 2020 Notes"
+date: 2020-05-05
+---
+
+ICLR is one of the most important conferences in machine learning, and
+as such, I was very excited to have the opportunity to volunteer and
+attend the first fully-virtual edition of the event. The whole content
+of the conference has been made [[https://iclr.cc/virtual_2020/index.html][publicly available]], only a few days
+after the end of the event!
+
+I would like to thank the [[https://iclr.cc/Conferences/2020/Committees][organizing committee]] for this incredible
+event, and the possibility to volunteer to help other
+participants[fn:volunteer].
+
+The many volunteers, the online-only nature of the event, and the low
+registration fees also allowed for what felt like a very diverse,
+inclusive event. Many graduate students and researchers from industry
+(like me), who do not generally have the time or the resources to
+travel to conferences like this, were able to attend, and make the
+exchanges richer.
+
+In this post, I will try to give my impressions on the event, and
+share the most interesting events and papers I saw.
+
+[fn:volunteer] To better organize the event, and help people navigate
+the various online tools, they brought in 500(!) volunteers, waved our
+registration fees, and asked us to do simple load-testing and tech
+support. This was a very generous offer, and felt very rewarding for
+us, as we could attend the conference, and give back to the
+organization a little bit.
+
+* The Format of the Virtual Conference
+
+As a result of global travel restrictions, the conference was made
+fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia,
+which is great for people who are often the target of restrictive visa
+policies in Northern American countries.
+
+The thing I appreciated most about the conference format was its
+emphasis on /asynchronous/ communication. Given how little time they
+had to plan the conference, they could have made all poster
+presentations via video-conference and call it a day. Instead, each
+poster had to record a 5-minute video summarising their
+research. Alongside each presentation, there was a dedicated
+Rocket.Chat channel[fn:rocketchat] where anyone could ask a question
+to the authors, or just show their appreciation for the work. This was
+a fantastic idea as it allowed any participant to interact with papers
+and authors at any time they please, which is especially important in
+a setting where people were spread all over the globe.
+
+There were also Zoom session where authors were available for direct,
+face-to-face discussions, allowing for more traditional
+conversations. But asking questions on the channel had also the
+advantage of keeping a track of all questions that were asked by other
+people. As such, I quickly acquired the habit of watching the video,
+looking at the chat to see the previous discussions (even if they
+happened in the middle of the night in my timezone!), and then
+skimming the paper or asking questions myself.
+
+All of these excellent ideas were implemented by an [[https://iclr.cc/virtual_2020/papers.html?filter=keywords][amazing website]],
+collecting all papers in a searchable, easy-to-use interface, and even
+a nice [[https://iclr.cc/virtual_2020/paper_vis.html][visualisation]] of papers as a point cloud!
+
+[fn:rocketchat] [[https://rocket.chat/][Rocket.Chat]] seems to be an [[https://github.com/RocketChat/Rocket.Chat][open-source]] alternative to
+Slack. Overall, the experience was great, and I appreciate the efforts
+of the organizers to use open source software instead of proprietary
+applications. I hope other conferences will do the same, and perhaps
+even avoid Zoom, because of recent privacy concerns (maybe try
+[[https://jitsi.org/][Jitsi]]?).
+
+* Speakers
+
+Overall, there were 8 speakers (two for each day of the main
+conference). They made a 40-minute presentation, and then there was a
+Q&A both via the chat and via Zoom. I only saw 4 of them, but I expect
+I will be watching the others in the near future.
+
+** Prof. Leslie Kaelbling, [[https://iclr.cc/virtual_2020/speaker_2.html][Doing for Our Robots What Nature Did For Us]]
+
+This talk was fascinating. It is about robotics, and especially how to
+design the "software" of our robots. We want to program a robot in a
+way that it could work the best it can over all possible domains it
+can encounter.
+
+
+
+* Workshops
+
+* Some Interesting Papers
+
+** Natural Language Processing
+
+*** Harwath et al., [[https://openreview.net/forum?id=B1elCp4KwH][Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech]]
+
+Humans can easily deconstruct all the information available in speech
+(meaning, language, emotion, speaker, etc.). However, this is very
+hard for machines. This paper explores the capacity of algorithms to
+reason about all the aspects of the signal, including visual cues.
+
+Their goal is to use spoken captions of images to train a predictive
+model.
+
+- speech signal: contains a lot of information (meaning, language,
+  emotion, speaker, environment)
+  - difficult to identify these different parts for an algo (although
+    easy for a human)
+- dominated by supervised ML
+  - automated speech recognition (ASR) = P(text | waveform)
+  - text-to-speech (TTS) = P(waveform | text, speaker)
+  - high sample complexity
+  - bad out-of-domain performance
+  - limited by annotation capability
+- human-like learning
+  - ability to jointly reason about all the aspects of the signal
+  - rapidly adapt to new speaker or noise conditions
+  - learn new words for a single example
+  - utilize unlabelled multimodal data
+- using visual grounding for self-supervision
+  - humans can leverage cross-modal correspondences to learn what
+    spoken words represent without requiring any text or symbolic
+    input whatsoever
+  - hypothesis: similar for computer algorithms?
+- goal: use spoken captions of images to train a predictive model
+  - learn a hierarchical structure of units
+  - learn the corresponding text, but also the transcription of the
+    spoken sounds, at a sub-word level
+- prefer models that learn a discrete tokenisation of the speech
+  - language has an intrinsically symbolic structure
+    - convey meaning with discrete words
+    - words are in turn composed of a finite set of speech sounds
+      (phones)
+  - model that can discover discrete representations for word and
+    phone-like units
+    - more interpretable
+    - able to do few-shot learning (learn a new word-like unit in
+      terms of known phone-like units)
+    - path towards learning compositional structure from continuous
+      signals
+- model for audio-visual grounding:
+  - NN for image
+  - NN for raw speech
+  - shared embedding space
+  - semantic supervision
+- preliminary studies
+  - lower layers features are correlated with phones
+  - higher lauer features are correlated with words
+- add a vector quantizing layer in the speech NN
+- hierarchy of quantization layers
+  - capture phones and words
+- do an ABX test to compare performance to speech-only models
+- conclusion
+  - novel linguistic unit learning paradigm using multimodal data
+    without text
+  - SOTA performance on learning phonetic and word-level units
+  - discovery of discreteness as a good inductive bias for semantic
+    task from speech
+
+** Reinforcement Learning
+
+** ML and Neural Network Theory
+

From 149d0a03003da795ac707f9b4b5a26f001a2a267 Mon Sep 17 00:00:00 2001
From: Dimitri Lozeve <dimitri@lozeve.com>
Date: Tue, 5 May 2020 10:26:19 +0200
Subject: [PATCH 2/6] Add notes on speakers

---
 _site/atom.xml                   | 14 ++++++++-
 _site/posts/iclr-2020-notes.html | 14 ++++++++-
 _site/rss.xml                    | 14 ++++++++-
 posts/iclr-2020-notes.org        | 52 +++++++++++++++++++++++++++++++-
 4 files changed, 90 insertions(+), 4 deletions(-)

diff --git a/_site/atom.xml b/_site/atom.xml
index bd3c5d0..508d787 100644
--- a/_site/atom.xml
+++ b/_site/atom.xml
@@ -36,7 +36,19 @@
 <h1 id="speakers">Speakers</h1>
 <p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
 <h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
-<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter.</p>
+<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter. I loved the discussion on how to describe the space of distributions over domains, from the point of view of the robot factory:</p>
+<ul>
+<li>The domain could be very narrow (e.g. playing a specific Atari game) or very broad and complex (performing a complex task in an open world).</li>
+<li>The factory could know in advance in which domain the robot will evolve, or have a lot of uncertainty around it.</li>
+</ul>
+<p>There are many ways to describe a policy (i.e. the software running in the robot’s head), and many ways to obtain them. If you are familiar with recent advances in reinforcement learning, this talk is a great occasion to take a step back, and review the relevant background ideas from engineering and control theory.</p>
+<p>Finally, the most important take-away from this talk is the importance of <em>abstractions</em>. Whatever the methods we use to program our robots, we still need a lot of human insights to give them good structural biases. There are many more insights, on the cost of experience, (hierarchical) planning, learning constraints, etc, so I strongly encourage you to watch the talk!</p>
+<h2 id="dr.-laurent-dinh-invertible-models-and-normalizing-flows">Dr. Laurent Dinh, <a href="https://iclr.cc/virtual_2020/speaker_4.html">Invertible Models and Normalizing Flows</a></h2>
+<p>This is a talk about an area of ML research I do not know very well, but very clearly presented. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
+<h2 id="profs.-yann-lecun-and-yoshua-bengio-reflections-from-the-turing-award-winners">Profs. Yann LeCun and Yoshua Bengio, <a href="https://iclr.cc/virtual_2020/speaker_7.html">Reflections from the Turing Award Winners</a></h2>
+<p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
+<h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
+<p>TODO</p>
 <h1 id="workshops">Workshops</h1>
 <h1 id="some-interesting-papers">Some Interesting Papers</h1>
 <h2 id="natural-language-processing">Natural Language Processing</h2>
diff --git a/_site/posts/iclr-2020-notes.html b/_site/posts/iclr-2020-notes.html
index 3281e1e..6472c9a 100644
--- a/_site/posts/iclr-2020-notes.html
+++ b/_site/posts/iclr-2020-notes.html
@@ -65,7 +65,19 @@
 <h1 id="speakers">Speakers</h1>
 <p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
 <h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
-<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter.</p>
+<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter. I loved the discussion on how to describe the space of distributions over domains, from the point of view of the robot factory:</p>
+<ul>
+<li>The domain could be very narrow (e.g. playing a specific Atari game) or very broad and complex (performing a complex task in an open world).</li>
+<li>The factory could know in advance in which domain the robot will evolve, or have a lot of uncertainty around it.</li>
+</ul>
+<p>There are many ways to describe a policy (i.e. the software running in the robot’s head), and many ways to obtain them. If you are familiar with recent advances in reinforcement learning, this talk is a great occasion to take a step back, and review the relevant background ideas from engineering and control theory.</p>
+<p>Finally, the most important take-away from this talk is the importance of <em>abstractions</em>. Whatever the methods we use to program our robots, we still need a lot of human insights to give them good structural biases. There are many more insights, on the cost of experience, (hierarchical) planning, learning constraints, etc, so I strongly encourage you to watch the talk!</p>
+<h2 id="dr.-laurent-dinh-invertible-models-and-normalizing-flows">Dr. Laurent Dinh, <a href="https://iclr.cc/virtual_2020/speaker_4.html">Invertible Models and Normalizing Flows</a></h2>
+<p>This is a talk about an area of ML research I do not know very well, but very clearly presented. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
+<h2 id="profs.-yann-lecun-and-yoshua-bengio-reflections-from-the-turing-award-winners">Profs. Yann LeCun and Yoshua Bengio, <a href="https://iclr.cc/virtual_2020/speaker_7.html">Reflections from the Turing Award Winners</a></h2>
+<p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
+<h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
+<p>TODO</p>
 <h1 id="workshops">Workshops</h1>
 <h1 id="some-interesting-papers">Some Interesting Papers</h1>
 <h2 id="natural-language-processing">Natural Language Processing</h2>
diff --git a/_site/rss.xml b/_site/rss.xml
index 9b0802f..65df8e7 100644
--- a/_site/rss.xml
+++ b/_site/rss.xml
@@ -32,7 +32,19 @@
 <h1 id="speakers">Speakers</h1>
 <p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
 <h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
-<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter.</p>
+<p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter. I loved the discussion on how to describe the space of distributions over domains, from the point of view of the robot factory:</p>
+<ul>
+<li>The domain could be very narrow (e.g. playing a specific Atari game) or very broad and complex (performing a complex task in an open world).</li>
+<li>The factory could know in advance in which domain the robot will evolve, or have a lot of uncertainty around it.</li>
+</ul>
+<p>There are many ways to describe a policy (i.e. the software running in the robot’s head), and many ways to obtain them. If you are familiar with recent advances in reinforcement learning, this talk is a great occasion to take a step back, and review the relevant background ideas from engineering and control theory.</p>
+<p>Finally, the most important take-away from this talk is the importance of <em>abstractions</em>. Whatever the methods we use to program our robots, we still need a lot of human insights to give them good structural biases. There are many more insights, on the cost of experience, (hierarchical) planning, learning constraints, etc, so I strongly encourage you to watch the talk!</p>
+<h2 id="dr.-laurent-dinh-invertible-models-and-normalizing-flows">Dr. Laurent Dinh, <a href="https://iclr.cc/virtual_2020/speaker_4.html">Invertible Models and Normalizing Flows</a></h2>
+<p>This is a talk about an area of ML research I do not know very well, but very clearly presented. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
+<h2 id="profs.-yann-lecun-and-yoshua-bengio-reflections-from-the-turing-award-winners">Profs. Yann LeCun and Yoshua Bengio, <a href="https://iclr.cc/virtual_2020/speaker_7.html">Reflections from the Turing Award Winners</a></h2>
+<p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
+<h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
+<p>TODO</p>
 <h1 id="workshops">Workshops</h1>
 <h1 id="some-interesting-papers">Some Interesting Papers</h1>
 <h2 id="natural-language-processing">Natural Language Processing</h2>
diff --git a/posts/iclr-2020-notes.org b/posts/iclr-2020-notes.org
index cd53eea..93c732e 100644
--- a/posts/iclr-2020-notes.org
+++ b/posts/iclr-2020-notes.org
@@ -81,9 +81,59 @@ I will be watching the others in the near future.
 This talk was fascinating. It is about robotics, and especially how to
 design the "software" of our robots. We want to program a robot in a
 way that it could work the best it can over all possible domains it
-can encounter.
+can encounter. I loved the discussion on how to describe the space of
+distributions over domains, from the point of view of the robot
+factory:
+- The domain could be very narrow (e.g. playing a specific Atari game)
+  or very broad and complex (performing a complex task in an open
+  world).
+- The factory could know in advance in which domain the robot will
+  evolve, or have a lot of uncertainty around it.
 
+There are many ways to describe a policy (i.e. the software running in
+the robot's head), and many ways to obtain them. If you are familiar
+with recent advances in reinforcement learning, this talk is a great
+occasion to take a step back, and review the relevant background ideas
+from engineering and control theory.
 
+Finally, the most important take-away from this talk is the importance
+of /abstractions/. Whatever the methods we use to program our robots,
+we still need a lot of human insights to give them good structural
+biases. There are many more insights, on the cost of experience,
+(hierarchical) planning, learning constraints, etc, so I strongly
+encourage you to watch the talk!
+
+** Dr. Laurent Dinh, [[https://iclr.cc/virtual_2020/speaker_4.html][Invertible Models and Normalizing Flows]]
+
+This is a talk about an area of ML research I do not know very well,
+but very clearly presented. I really like the approach of teaching a
+set of methods from a "historical", personal point of view. Laurent
+Dinh shows us how he arrived at this topic, what he finds interesting,
+in a very personal and relatable manner. This has the double advantage
+of introducing us to a topic that he is passionate about, while also
+giving us a glimpse of a researcher's process, without hiding the
+momentary disillusions and disappointments, but emphasising the great
+achievements. Normalizing flows are also very interesting because it
+is grounded in strong theoretical results, that brings together a lot
+of different methods.
+
+** Profs. Yann LeCun and Yoshua Bengio, [[https://iclr.cc/virtual_2020/speaker_7.html][Reflections from the Turing Award Winners]]
+
+This talk was very interesting, and yet felt very familiar, as if I
+already saw a very similar one elsewhere. Especially for Yann LeCun,
+who clearly reuses the same slides for many presentations at various
+events. They both came back to their favourite subjects:
+self-supervised learning for Yann LeCun, and system 1/system 2 for
+Yoshua Bengio. All in all, they are very good speakers, and their
+presentations are always insightful. Yann LeCun gives a lot of
+references on recent technical advances, which is great if you want to
+go deeper in the approaches he recommends. Yoshua Bengio is also very
+good at broadening the debate around deep learning, and introducing
+very important concepts from cognitive science.
+
+** Prof. Michael I. Jordan, [[https://iclr.cc/virtual_2020/speaker_8.html][The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives]]
+
+TODO
 
 * Workshops
 

From 41b844abe01d895efbc4fe901fa85bc53145830d Mon Sep 17 00:00:00 2001
From: Dimitri Lozeve <dimitri@lozeve.com>
Date: Tue, 5 May 2020 10:29:54 +0200
Subject: [PATCH 3/6] Prepare papers sections

---
 _site/atom.xml                   | 74 +-------------------------------
 _site/posts/iclr-2020-notes.html | 74 +-------------------------------
 _site/rss.xml                    | 74 +-------------------------------
 posts/iclr-2020-notes.org        | 68 +----------------------------
 4 files changed, 4 insertions(+), 286 deletions(-)

diff --git a/_site/atom.xml b/_site/atom.xml
index 508d787..6a15a29 100644
--- a/_site/atom.xml
+++ b/_site/atom.xml
@@ -49,83 +49,11 @@
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
 <h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
 <p>TODO</p>
-<h1 id="workshops">Workshops</h1>
 <h1 id="some-interesting-papers">Some Interesting Papers</h1>
 <h2 id="natural-language-processing">Natural Language Processing</h2>
-<h3 id="harwath-et-al.-learning-hierarchical-discrete-linguistic-units-from-visually-grounded-speech">Harwath et al., <a href="https://openreview.net/forum?id=B1elCp4KwH">Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech</a></h3>
-<p>Humans can easily deconstruct all the information available in speech (meaning, language, emotion, speaker, etc.). However, this is very hard for machines. This paper explores the capacity of algorithms to reason about all the aspects of the signal, including visual cues.</p>
-<p>Their goal is to use spoken captions of images to train a predictive model.</p>
-<ul>
-<li>speech signal: contains a lot of information (meaning, language, emotion, speaker, environment)
-<ul>
-<li>difficult to identify these different parts for an algo (although easy for a human)</li>
-</ul></li>
-<li>dominated by supervised ML
-<ul>
-<li>automated speech recognition (ASR) = P(text | waveform)</li>
-<li>text-to-speech (TTS) = P(waveform | text, speaker)</li>
-<li>high sample complexity</li>
-<li>bad out-of-domain performance</li>
-<li>limited by annotation capability</li>
-</ul></li>
-<li>human-like learning
-<ul>
-<li>ability to jointly reason about all the aspects of the signal</li>
-<li>rapidly adapt to new speaker or noise conditions</li>
-<li>learn new words for a single example</li>
-<li>utilize unlabelled multimodal data</li>
-</ul></li>
-<li>using visual grounding for self-supervision
-<ul>
-<li>humans can leverage cross-modal correspondences to learn what spoken words represent without requiring any text or symbolic input whatsoever</li>
-<li>hypothesis: similar for computer algorithms?</li>
-</ul></li>
-<li>goal: use spoken captions of images to train a predictive model
-<ul>
-<li>learn a hierarchical structure of units</li>
-<li>learn the corresponding text, but also the transcription of the spoken sounds, at a sub-word level</li>
-</ul></li>
-<li>prefer models that learn a discrete tokenisation of the speech
-<ul>
-<li>language has an intrinsically symbolic structure
-<ul>
-<li>convey meaning with discrete words</li>
-<li>words are in turn composed of a finite set of speech sounds (phones)</li>
-</ul></li>
-<li>model that can discover discrete representations for word and phone-like units
-<ul>
-<li>more interpretable</li>
-<li>able to do few-shot learning (learn a new word-like unit in terms of known phone-like units)</li>
-<li>path towards learning compositional structure from continuous signals</li>
-</ul></li>
-</ul></li>
-<li>model for audio-visual grounding:
-<ul>
-<li>NN for image</li>
-<li>NN for raw speech</li>
-<li>shared embedding space</li>
-<li>semantic supervision</li>
-</ul></li>
-<li>preliminary studies
-<ul>
-<li>lower layers features are correlated with phones</li>
-<li>higher lauer features are correlated with words</li>
-</ul></li>
-<li>add a vector quantizing layer in the speech NN</li>
-<li>hierarchy of quantization layers
-<ul>
-<li>capture phones and words</li>
-</ul></li>
-<li>do an ABX test to compare performance to speech-only models</li>
-<li>conclusion
-<ul>
-<li>novel linguistic unit learning paradigm using multimodal data without text</li>
-<li>SOTA performance on learning phonetic and word-level units</li>
-<li>discovery of discreteness as a good inductive bias for semantic task from speech</li>
-</ul></li>
-</ul>
 <h2 id="reinforcement-learning">Reinforcement Learning</h2>
 <h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
+<h1 id="workshops">Workshops</h1>
     </section>
 </article>
 ]]></summary>
diff --git a/_site/posts/iclr-2020-notes.html b/_site/posts/iclr-2020-notes.html
index 6472c9a..ada5049 100644
--- a/_site/posts/iclr-2020-notes.html
+++ b/_site/posts/iclr-2020-notes.html
@@ -78,83 +78,11 @@
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
 <h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
 <p>TODO</p>
-<h1 id="workshops">Workshops</h1>
 <h1 id="some-interesting-papers">Some Interesting Papers</h1>
 <h2 id="natural-language-processing">Natural Language Processing</h2>
-<h3 id="harwath-et-al.-learning-hierarchical-discrete-linguistic-units-from-visually-grounded-speech">Harwath et al., <a href="https://openreview.net/forum?id=B1elCp4KwH">Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech</a></h3>
-<p>Humans can easily deconstruct all the information available in speech (meaning, language, emotion, speaker, etc.). However, this is very hard for machines. This paper explores the capacity of algorithms to reason about all the aspects of the signal, including visual cues.</p>
-<p>Their goal is to use spoken captions of images to train a predictive model.</p>
-<ul>
-<li>speech signal: contains a lot of information (meaning, language, emotion, speaker, environment)
-<ul>
-<li>difficult to identify these different parts for an algo (although easy for a human)</li>
-</ul></li>
-<li>dominated by supervised ML
-<ul>
-<li>automated speech recognition (ASR) = P(text | waveform)</li>
-<li>text-to-speech (TTS) = P(waveform | text, speaker)</li>
-<li>high sample complexity</li>
-<li>bad out-of-domain performance</li>
-<li>limited by annotation capability</li>
-</ul></li>
-<li>human-like learning
-<ul>
-<li>ability to jointly reason about all the aspects of the signal</li>
-<li>rapidly adapt to new speaker or noise conditions</li>
-<li>learn new words for a single example</li>
-<li>utilize unlabelled multimodal data</li>
-</ul></li>
-<li>using visual grounding for self-supervision
-<ul>
-<li>humans can leverage cross-modal correspondences to learn what spoken words represent without requiring any text or symbolic input whatsoever</li>
-<li>hypothesis: similar for computer algorithms?</li>
-</ul></li>
-<li>goal: use spoken captions of images to train a predictive model
-<ul>
-<li>learn a hierarchical structure of units</li>
-<li>learn the corresponding text, but also the transcription of the spoken sounds, at a sub-word level</li>
-</ul></li>
-<li>prefer models that learn a discrete tokenisation of the speech
-<ul>
-<li>language has an intrinsically symbolic structure
-<ul>
-<li>convey meaning with discrete words</li>
-<li>words are in turn composed of a finite set of speech sounds (phones)</li>
-</ul></li>
-<li>model that can discover discrete representations for word and phone-like units
-<ul>
-<li>more interpretable</li>
-<li>able to do few-shot learning (learn a new word-like unit in terms of known phone-like units)</li>
-<li>path towards learning compositional structure from continuous signals</li>
-</ul></li>
-</ul></li>
-<li>model for audio-visual grounding:
-<ul>
-<li>NN for image</li>
-<li>NN for raw speech</li>
-<li>shared embedding space</li>
-<li>semantic supervision</li>
-</ul></li>
-<li>preliminary studies
-<ul>
-<li>lower layers features are correlated with phones</li>
-<li>higher lauer features are correlated with words</li>
-</ul></li>
-<li>add a vector quantizing layer in the speech NN</li>
-<li>hierarchy of quantization layers
-<ul>
-<li>capture phones and words</li>
-</ul></li>
-<li>do an ABX test to compare performance to speech-only models</li>
-<li>conclusion
-<ul>
-<li>novel linguistic unit learning paradigm using multimodal data without text</li>
-<li>SOTA performance on learning phonetic and word-level units</li>
-<li>discovery of discreteness as a good inductive bias for semantic task from speech</li>
-</ul></li>
-</ul>
 <h2 id="reinforcement-learning">Reinforcement Learning</h2>
 <h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
+<h1 id="workshops">Workshops</h1>
     </section>
 </article>
 
diff --git a/_site/rss.xml b/_site/rss.xml
index 65df8e7..2da456d 100644
--- a/_site/rss.xml
+++ b/_site/rss.xml
@@ -45,83 +45,11 @@
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
 <h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
 <p>TODO</p>
-<h1 id="workshops">Workshops</h1>
 <h1 id="some-interesting-papers">Some Interesting Papers</h1>
 <h2 id="natural-language-processing">Natural Language Processing</h2>
-<h3 id="harwath-et-al.-learning-hierarchical-discrete-linguistic-units-from-visually-grounded-speech">Harwath et al., <a href="https://openreview.net/forum?id=B1elCp4KwH">Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech</a></h3>
-<p>Humans can easily deconstruct all the information available in speech (meaning, language, emotion, speaker, etc.). However, this is very hard for machines. This paper explores the capacity of algorithms to reason about all the aspects of the signal, including visual cues.</p>
-<p>Their goal is to use spoken captions of images to train a predictive model.</p>
-<ul>
-<li>speech signal: contains a lot of information (meaning, language, emotion, speaker, environment)
-<ul>
-<li>difficult to identify these different parts for an algo (although easy for a human)</li>
-</ul></li>
-<li>dominated by supervised ML
-<ul>
-<li>automated speech recognition (ASR) = P(text | waveform)</li>
-<li>text-to-speech (TTS) = P(waveform | text, speaker)</li>
-<li>high sample complexity</li>
-<li>bad out-of-domain performance</li>
-<li>limited by annotation capability</li>
-</ul></li>
-<li>human-like learning
-<ul>
-<li>ability to jointly reason about all the aspects of the signal</li>
-<li>rapidly adapt to new speaker or noise conditions</li>
-<li>learn new words for a single example</li>
-<li>utilize unlabelled multimodal data</li>
-</ul></li>
-<li>using visual grounding for self-supervision
-<ul>
-<li>humans can leverage cross-modal correspondences to learn what spoken words represent without requiring any text or symbolic input whatsoever</li>
-<li>hypothesis: similar for computer algorithms?</li>
-</ul></li>
-<li>goal: use spoken captions of images to train a predictive model
-<ul>
-<li>learn a hierarchical structure of units</li>
-<li>learn the corresponding text, but also the transcription of the spoken sounds, at a sub-word level</li>
-</ul></li>
-<li>prefer models that learn a discrete tokenisation of the speech
-<ul>
-<li>language has an intrinsically symbolic structure
-<ul>
-<li>convey meaning with discrete words</li>
-<li>words are in turn composed of a finite set of speech sounds (phones)</li>
-</ul></li>
-<li>model that can discover discrete representations for word and phone-like units
-<ul>
-<li>more interpretable</li>
-<li>able to do few-shot learning (learn a new word-like unit in terms of known phone-like units)</li>
-<li>path towards learning compositional structure from continuous signals</li>
-</ul></li>
-</ul></li>
-<li>model for audio-visual grounding:
-<ul>
-<li>NN for image</li>
-<li>NN for raw speech</li>
-<li>shared embedding space</li>
-<li>semantic supervision</li>
-</ul></li>
-<li>preliminary studies
-<ul>
-<li>lower layers features are correlated with phones</li>
-<li>higher lauer features are correlated with words</li>
-</ul></li>
-<li>add a vector quantizing layer in the speech NN</li>
-<li>hierarchy of quantization layers
-<ul>
-<li>capture phones and words</li>
-</ul></li>
-<li>do an ABX test to compare performance to speech-only models</li>
-<li>conclusion
-<ul>
-<li>novel linguistic unit learning paradigm using multimodal data without text</li>
-<li>SOTA performance on learning phonetic and word-level units</li>
-<li>discovery of discreteness as a good inductive bias for semantic task from speech</li>
-</ul></li>
-</ul>
 <h2 id="reinforcement-learning">Reinforcement Learning</h2>
 <h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
+<h1 id="workshops">Workshops</h1>
     </section>
 </article>
 ]]></description>
diff --git a/posts/iclr-2020-notes.org b/posts/iclr-2020-notes.org
index 93c732e..39a4701 100644
--- a/posts/iclr-2020-notes.org
+++ b/posts/iclr-2020-notes.org
@@ -135,78 +135,12 @@ very important concepts from cognitive science.
 
 TODO
 
-* Workshops
-
 * Some Interesting Papers
 
 ** Natural Language Processing
 
-*** Harwath et al., [[https://openreview.net/forum?id=B1elCp4KwH][Learning Hierarchical Discrete Linguistic Units from Visually-Grounded Speech]]
-
-Humans can easily deconstruct all the information available in speech
-(meaning, language, emotion, speaker, etc.). However, this is very
-hard for machines. This paper explores the capacity of algorithms to
-reason about all the aspects of the signal, including visual cues.
-
-Their goal is to use spoken captions of images to train a predictive
-model.
-
-- speech signal: contains a lot of information (meaning, language,
-  emotion, speaker, environment)
-  - difficult to identify these different parts for an algo (although
-    easy for a human)
-- dominated by supervised ML
-  - automated speech recognition (ASR) = P(text | waveform)
-  - text-to-speech (TTS) = P(waveform | text, speaker)
-  - high sample complexity
-  - bad out-of-domain performance
-  - limited by annotation capability
-- human-like learning
-  - ability to jointly reason about all the aspects of the signal
-  - rapidly adapt to new speaker or noise conditions
-  - learn new words for a single example
-  - utilize unlabelled multimodal data
-- using visual grounding for self-supervision
-  - humans can leverage cross-modal correspondences to learn what
-    spoken words represent without requiring any text or symbolic
-    input whatsoever
-  - hypothesis: similar for computer algorithms?
-- goal: use spoken captions of images to train a predictive model
-  - learn a hierarchical structure of units
-  - learn the corresponding text, but also the transcription of the
-    spoken sounds, at a sub-word level
-- prefer models that learn a discrete tokenisation of the speech
-  - language has an intrinsically symbolic structure
-    - convey meaning with discrete words
-    - words are in turn composed of a finite set of speech sounds
-      (phones)
-  - model that can discover discrete representations for word and
-    phone-like units
-    - more interpretable
-    - able to do few-shot learning (learn a new word-like unit in
-      terms of known phone-like units)
-    - path towards learning compositional structure from continuous
-      signals
-- model for audio-visual grounding:
-  - NN for image
-  - NN for raw speech
-  - shared embedding space
-  - semantic supervision
-- preliminary studies
-  - lower layers features are correlated with phones
-  - higher lauer features are correlated with words
-- add a vector quantizing layer in the speech NN
-- hierarchy of quantization layers
-  - capture phones and words
-- do an ABX test to compare performance to speech-only models
-- conclusion
-  - novel linguistic unit learning paradigm using multimodal data
-    without text
-  - SOTA performance on learning phonetic and word-level units
-  - discovery of discreteness as a good inductive bias for semantic
-    task from speech
-
 ** Reinforcement Learning
 
 ** ML and Neural Network Theory
 
+* Workshops

From 6f50c32ed1ef58409ea58bdf228b3685537457e8 Mon Sep 17 00:00:00 2001
From: Dimitri Lozeve <dimitri@lozeve.com>
Date: Tue, 5 May 2020 12:00:17 +0200
Subject: [PATCH 4/6] Fix stylesheet for correct footnote numbering

---
 _site/css/default.css | 2 +-
 css/default.css       | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/_site/css/default.css b/_site/css/default.css
index 5cd7836..5a228c5 100644
--- a/_site/css/default.css
+++ b/_site/css/default.css
@@ -1 +1 @@
-body{background:#fafafa;counter-reset:theorem;counter-reset:definition}header{margin:1em 0 2em 0}header nav{width:87.5%}header nav a{display:inline-block;font-size:1.4rem;font-weight:bold;text-decoration:none!important;margin:0 0.5rem;padding:0 0}header nav a:first-child{margin-left:0}header nav a:last-child{margin-right:0}footer{margin-top:3rem;padding:.5rem 0;border-top:0.1rem solid #000;color:#555}article .header{font-style:italic;color:#555}.definition,.proposition,.theorem{display:block;border-left:2px solid #808080;padding-left:1rem;margin-top:1rem;margin-bottom:1rem}.proposition,.theorem{counter-increment:theorem;font-style:italic}.definition p:first-child,.proposition p:first-child,.theorem p:first-child,.proof p:first-child{display:inline}.definition p,.proposition p,.theorem p,.proof p{margin-top:0.4rem;margin-bottom:0.4rem;padding-left:1rem}.theorem:before{content:"Theorem " counter(theorem) ".";font-weight:bold;font-style:normal}.proposition:before{content:"Proposition " counter(theorem) ".";font-weight:bold;font-style:normal}.definition:before{counter-increment:definition;content:"Definition " counter(definition) ".";font-weight:bold;font-style:normal}.proof:before{content:"Proof.";font-style:italic}.proof:after{content:"\220E";float:right}
\ No newline at end of file
+body{background:#fafafa;counter-reset:theorem;counter-reset:definition;counter-reset:sidenote-counter}header{margin:1em 0 2em 0}header nav{width:87.5%}header nav a{display:inline-block;font-size:1.4rem;font-weight:bold;text-decoration:none!important;margin:0 0.5rem;padding:0 0}header nav a:first-child{margin-left:0}header nav a:last-child{margin-right:0}footer{margin-top:3rem;padding:.5rem 0;border-top:0.1rem solid #000;color:#555}article .header{font-style:italic;color:#555}.definition,.proposition,.theorem{display:block;border-left:2px solid #808080;padding-left:1rem;margin-top:1rem;margin-bottom:1rem}.proposition,.theorem{counter-increment:theorem;font-style:italic}.definition p:first-child,.proposition p:first-child,.theorem p:first-child,.proof p:first-child{display:inline}.definition p,.proposition p,.theorem p,.proof p{margin-top:0.4rem;margin-bottom:0.4rem;padding-left:1rem}.theorem:before{content:"Theorem " counter(theorem) ".";font-weight:bold;font-style:normal}.proposition:before{content:"Proposition " counter(theorem) ".";font-weight:bold;font-style:normal}.definition:before{counter-increment:definition;content:"Definition " counter(definition) ".";font-weight:bold;font-style:normal}.proof:before{content:"Proof.";font-style:italic}.proof:after{content:"\220E";float:right}
\ No newline at end of file
diff --git a/css/default.css b/css/default.css
index 4b77227..6fc94b7 100644
--- a/css/default.css
+++ b/css/default.css
@@ -2,6 +2,7 @@ body {
     background: #fafafa;
     counter-reset: theorem;
     counter-reset: definition;
+    counter-reset: sidenote-counter; /* Hack, otherwise we break Tufte CSS */
 }
 
 header {

From 17b25af9c5b58cbe6e690193bdacb648cc44bdb8 Mon Sep 17 00:00:00 2001
From: Dimitri Lozeve <dimitri@lozeve.com>
Date: Tue, 5 May 2020 12:12:00 +0200
Subject: [PATCH 5/6] Add workshop notes

---
 _site/archive.html               |  2 +-
 _site/atom.xml                   | 21 ++++++---
 _site/index.html                 |  2 +-
 _site/posts/iclr-2020-notes.html | 25 +++++++----
 _site/rss.xml                    | 21 ++++++---
 posts/iclr-2020-notes.org        | 74 +++++++++++++++++++++++++++-----
 6 files changed, 113 insertions(+), 32 deletions(-)

diff --git a/_site/archive.html b/_site/archive.html
index fefe245..370c561 100644
--- a/_site/archive.html
+++ b/_site/archive.html
@@ -48,7 +48,7 @@
 <ul>
     
         <li>
-            <a href="./posts/iclr-2020-notes.html">ICLR 2020 Notes</a> - May  5, 2020
+            <a href="./posts/iclr-2020-notes.html">ICLR 2020 Notes: Speakers and Workshops</a> - May  5, 2020
         </li>
     
         <li>
diff --git a/_site/atom.xml b/_site/atom.xml
index 6a15a29..551993a 100644
--- a/_site/atom.xml
+++ b/_site/atom.xml
@@ -10,7 +10,7 @@
     </author>
     <updated>2020-05-05T00:00:00Z</updated>
     <entry>
-    <title>ICLR 2020 Notes</title>
+    <title>ICLR 2020 Notes: Speakers and Workshops</title>
     <link href="https://www.lozeve.com/posts/iclr-2020-notes.html" />
     <id>https://www.lozeve.com/posts/iclr-2020-notes.html</id>
     <published>2020-05-05T00:00:00Z</published>
@@ -28,7 +28,9 @@
 <p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
 <h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
 <p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
-<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
+<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote">The videos are streamed using <a href="https://library.slideslive.com/">SlidesLive</a>, which is a great solution for synchronising videos and slides. It is very comfortable to navigate through the slides and synchronising the video to the slides and vice-versa. As a result, SlidesLive also has a very nice library of talks, including major conferences. This is much better than browsing YouTube randomly.<br />
+<br />
+</span></span> summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-3" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-3" class="margin-toggle"/><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
 <br />
 </span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
 <p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
@@ -49,11 +51,18 @@
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
 <h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
 <p>TODO</p>
-<h1 id="some-interesting-papers">Some Interesting Papers</h1>
-<h2 id="natural-language-processing">Natural Language Processing</h2>
-<h2 id="reinforcement-learning">Reinforcement Learning</h2>
-<h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
 <h1 id="workshops">Workshops</h1>
+<p>On Sunday, there were <a href="https://iclr.cc/virtual_2020/workshops.html">15 different workshops</a>. All of them were recorded, and are available on the website. As always, unfortunately, there are too many interesting things to watch everything, but I saw bits and pieces of different workshops.</p>
+<h2 id="beyond-tabula-rasa-in-reinforcement-learning-agents-that-remember-adapt-and-generalize"><a href="https://iclr.cc/virtual_2020/workshops_12.html">Beyond ‘tabula rasa’ in reinforcement learning: agents that remember, adapt, and generalize</a></h2>
+<p>A lot of pretty advanced talks about RL. The general theme was meta-learning, aka “learning to learn”. This is a very active area of research, which goes way beyond classical RL theory, and offer many interesting avenues to adjacent fields (both inside ML and outside, especially cognitive science). The <a href="http://www.betr-rl.ml/2020/abs/101/">first talk</a>, by Martha White, about inductive biases, was a very interesting and approachable introduction to the problems and challenges of the field. There was also a panel with Jürgen Schmidhuber. We hear a lot about him from the various controversies, but it’s nice to see him talking about research and future developments in RL.</p>
+<h2 id="causal-learning-for-decision-making"><a href="https://iclr.cc/virtual_2020/workshops_14.html">Causal Learning For Decision Making</a></h2>
+<p>Ever since I read Judea Pearl’s <a href="https://www.goodreads.com/book/show/36204378-the-book-of-why"><em>The Book of Why</em></a> on causality, I have been interested in how we can incorporate causality reasoning in machine learning. This is a complex topic, and I’m not sure yet that it is a complete revolution as Judea Pearl likes to portray it, but it nevertheless introduces a lot of new fascinating ideas. Yoshua Bengio gave an interesting talk<span><label for="sn-4" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-4" class="margin-toggle"/><span class="sidenote">You can find it at 4:45:20 in the <a href="https://slideslive.com/38926830/workshop-on-causal-learning-for-decision-making">livestream</a> of the workshop.<br />
+<br />
+</span></span> (even though very similar to his keynote talk) on causal priors for deep learning.</p>
+<h2 id="bridging-ai-and-cognitive-science"><a href="https://iclr.cc/virtual_2020/workshops_4.html">Bridging AI and Cognitive Science</a></h2>
+<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the rights approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
+<h2 id="integration-of-deep-neural-models-and-differential-equations"><a href="https://iclr.cc/virtual_2020/workshops_5.html">Integration of Deep Neural Models and Differential Equations</a></h2>
+<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find some time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
     </section>
 </article>
 ]]></summary>
diff --git a/_site/index.html b/_site/index.html
index 87f633e..db75642 100644
--- a/_site/index.html
+++ b/_site/index.html
@@ -74,7 +74,7 @@ public key: RWQ6uexORp8f7USHA7nX9lFfltaCA9x6aBV06MvgiGjUt6BVf6McyD26
 <ul>
     
         <li>
-            <a href="./posts/iclr-2020-notes.html">ICLR 2020 Notes</a> - May  5, 2020
+            <a href="./posts/iclr-2020-notes.html">ICLR 2020 Notes: Speakers and Workshops</a> - May  5, 2020
         </li>
     
         <li>
diff --git a/_site/posts/iclr-2020-notes.html b/_site/posts/iclr-2020-notes.html
index ada5049..698f893 100644
--- a/_site/posts/iclr-2020-notes.html
+++ b/_site/posts/iclr-2020-notes.html
@@ -4,9 +4,9 @@
     <meta charset="utf-8">
     <meta http-equiv="x-ua-compatible" content="ie=edge">
     <meta name="viewport" content="width=device-width, initial-scale=1, user-scalable=yes">
-    <meta name="description" content="Dimitri Lozeve's blog: ICLR 2020 Notes">
+    <meta name="description" content="Dimitri Lozeve's blog: ICLR 2020 Notes: Speakers and Workshops">
     
-    <title>Dimitri Lozeve - ICLR 2020 Notes</title>
+    <title>Dimitri Lozeve - ICLR 2020 Notes: Speakers and Workshops</title>
     <link rel="stylesheet" href="../css/tufte.css" />
     <link rel="stylesheet" href="../css/pandoc.css" />
     <link rel="stylesheet" href="../css/default.css" />
@@ -33,7 +33,7 @@
 	  <a href="../contact.html">Contact</a>
 	</nav>
 
-	<h1 class="title">ICLR 2020 Notes</h1>
+	<h1 class="title">ICLR 2020 Notes: Speakers and Workshops</h1>
 	
 	
 	<p class="byline">May  5, 2020</p>
@@ -57,7 +57,9 @@
 <p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
 <h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
 <p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
-<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle" /><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
+<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle" /><span class="sidenote">The videos are streamed using <a href="https://library.slideslive.com/">SlidesLive</a>, which is a great solution for synchronising videos and slides. It is very comfortable to navigate through the slides and synchronising the video to the slides and vice-versa. As a result, SlidesLive also has a very nice library of talks, including major conferences. This is much better than browsing YouTube randomly.<br />
+<br />
+</span></span> summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-3" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-3" class="margin-toggle" /><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
 <br />
 </span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
 <p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
@@ -78,11 +80,18 @@
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
 <h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
 <p>TODO</p>
-<h1 id="some-interesting-papers">Some Interesting Papers</h1>
-<h2 id="natural-language-processing">Natural Language Processing</h2>
-<h2 id="reinforcement-learning">Reinforcement Learning</h2>
-<h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
 <h1 id="workshops">Workshops</h1>
+<p>On Sunday, there were <a href="https://iclr.cc/virtual_2020/workshops.html">15 different workshops</a>. All of them were recorded, and are available on the website. As always, unfortunately, there are too many interesting things to watch everything, but I saw bits and pieces of different workshops.</p>
+<h2 id="beyond-tabula-rasa-in-reinforcement-learning-agents-that-remember-adapt-and-generalize"><a href="https://iclr.cc/virtual_2020/workshops_12.html">Beyond ‘tabula rasa’ in reinforcement learning: agents that remember, adapt, and generalize</a></h2>
+<p>A lot of pretty advanced talks about RL. The general theme was meta-learning, aka “learning to learn”. This is a very active area of research, which goes way beyond classical RL theory, and offer many interesting avenues to adjacent fields (both inside ML and outside, especially cognitive science). The <a href="http://www.betr-rl.ml/2020/abs/101/">first talk</a>, by Martha White, about inductive biases, was a very interesting and approachable introduction to the problems and challenges of the field. There was also a panel with Jürgen Schmidhuber. We hear a lot about him from the various controversies, but it’s nice to see him talking about research and future developments in RL.</p>
+<h2 id="causal-learning-for-decision-making"><a href="https://iclr.cc/virtual_2020/workshops_14.html">Causal Learning For Decision Making</a></h2>
+<p>Ever since I read Judea Pearl’s <a href="https://www.goodreads.com/book/show/36204378-the-book-of-why"><em>The Book of Why</em></a> on causality, I have been interested in how we can incorporate causality reasoning in machine learning. This is a complex topic, and I’m not sure yet that it is a complete revolution as Judea Pearl likes to portray it, but it nevertheless introduces a lot of new fascinating ideas. Yoshua Bengio gave an interesting talk<span><label for="sn-4" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-4" class="margin-toggle" /><span class="sidenote">You can find it at 4:45:20 in the <a href="https://slideslive.com/38926830/workshop-on-causal-learning-for-decision-making">livestream</a> of the workshop.<br />
+<br />
+</span></span> (even though very similar to his keynote talk) on causal priors for deep learning.</p>
+<h2 id="bridging-ai-and-cognitive-science"><a href="https://iclr.cc/virtual_2020/workshops_4.html">Bridging AI and Cognitive Science</a></h2>
+<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the rights approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
+<h2 id="integration-of-deep-neural-models-and-differential-equations"><a href="https://iclr.cc/virtual_2020/workshops_5.html">Integration of Deep Neural Models and Differential Equations</a></h2>
+<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find some time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
     </section>
 </article>
 
diff --git a/_site/rss.xml b/_site/rss.xml
index 2da456d..aa71ca8 100644
--- a/_site/rss.xml
+++ b/_site/rss.xml
@@ -9,7 +9,7 @@
                    type="application/rss+xml" />
         <lastBuildDate>Tue, 05 May 2020 00:00:00 UT</lastBuildDate>
         <item>
-    <title>ICLR 2020 Notes</title>
+    <title>ICLR 2020 Notes: Speakers and Workshops</title>
     <link>https://www.lozeve.com/posts/iclr-2020-notes.html</link>
     <description><![CDATA[<article>
     <section class="header">
@@ -24,7 +24,9 @@
 <p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
 <h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
 <p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
-<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
+<p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote">The videos are streamed using <a href="https://library.slideslive.com/">SlidesLive</a>, which is a great solution for synchronising videos and slides. It is very comfortable to navigate through the slides and synchronising the video to the slides and vice-versa. As a result, SlidesLive also has a very nice library of talks, including major conferences. This is much better than browsing YouTube randomly.<br />
+<br />
+</span></span> summarising their research. Alongside each presentation, there was a dedicated Rocket.Chat channel<span><label for="sn-3" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-3" class="margin-toggle"/><span class="sidenote"><a href="https://rocket.chat/">Rocket.Chat</a> seems to be an <a href="https://github.com/RocketChat/Rocket.Chat">open-source</a> alternative to Slack. Overall, the experience was great, and I appreciate the efforts of the organizers to use open source software instead of proprietary applications. I hope other conferences will do the same, and perhaps even avoid Zoom, because of recent privacy concerns (maybe try <a href="https://jitsi.org/">Jitsi</a>?).<br />
 <br />
 </span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
 <p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
@@ -45,11 +47,18 @@
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
 <h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
 <p>TODO</p>
-<h1 id="some-interesting-papers">Some Interesting Papers</h1>
-<h2 id="natural-language-processing">Natural Language Processing</h2>
-<h2 id="reinforcement-learning">Reinforcement Learning</h2>
-<h2 id="ml-and-neural-network-theory">ML and Neural Network Theory</h2>
 <h1 id="workshops">Workshops</h1>
+<p>On Sunday, there were <a href="https://iclr.cc/virtual_2020/workshops.html">15 different workshops</a>. All of them were recorded, and are available on the website. As always, unfortunately, there are too many interesting things to watch everything, but I saw bits and pieces of different workshops.</p>
+<h2 id="beyond-tabula-rasa-in-reinforcement-learning-agents-that-remember-adapt-and-generalize"><a href="https://iclr.cc/virtual_2020/workshops_12.html">Beyond ‘tabula rasa’ in reinforcement learning: agents that remember, adapt, and generalize</a></h2>
+<p>A lot of pretty advanced talks about RL. The general theme was meta-learning, aka “learning to learn”. This is a very active area of research, which goes way beyond classical RL theory, and offer many interesting avenues to adjacent fields (both inside ML and outside, especially cognitive science). The <a href="http://www.betr-rl.ml/2020/abs/101/">first talk</a>, by Martha White, about inductive biases, was a very interesting and approachable introduction to the problems and challenges of the field. There was also a panel with Jürgen Schmidhuber. We hear a lot about him from the various controversies, but it’s nice to see him talking about research and future developments in RL.</p>
+<h2 id="causal-learning-for-decision-making"><a href="https://iclr.cc/virtual_2020/workshops_14.html">Causal Learning For Decision Making</a></h2>
+<p>Ever since I read Judea Pearl’s <a href="https://www.goodreads.com/book/show/36204378-the-book-of-why"><em>The Book of Why</em></a> on causality, I have been interested in how we can incorporate causality reasoning in machine learning. This is a complex topic, and I’m not sure yet that it is a complete revolution as Judea Pearl likes to portray it, but it nevertheless introduces a lot of new fascinating ideas. Yoshua Bengio gave an interesting talk<span><label for="sn-4" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-4" class="margin-toggle"/><span class="sidenote">You can find it at 4:45:20 in the <a href="https://slideslive.com/38926830/workshop-on-causal-learning-for-decision-making">livestream</a> of the workshop.<br />
+<br />
+</span></span> (even though very similar to his keynote talk) on causal priors for deep learning.</p>
+<h2 id="bridging-ai-and-cognitive-science"><a href="https://iclr.cc/virtual_2020/workshops_4.html">Bridging AI and Cognitive Science</a></h2>
+<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the rights approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
+<h2 id="integration-of-deep-neural-models-and-differential-equations"><a href="https://iclr.cc/virtual_2020/workshops_5.html">Integration of Deep Neural Models and Differential Equations</a></h2>
+<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find some time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
     </section>
 </article>
 ]]></description>
diff --git a/posts/iclr-2020-notes.org b/posts/iclr-2020-notes.org
index 39a4701..5417b6f 100644
--- a/posts/iclr-2020-notes.org
+++ b/posts/iclr-2020-notes.org
@@ -1,5 +1,5 @@
 ---
-title: "ICLR 2020 Notes"
+title: "ICLR 2020 Notes: Speakers and Workshops"
 date: 2020-05-05
 ---
 
@@ -41,7 +41,7 @@ The thing I appreciated most about the conference format was its
 emphasis on /asynchronous/ communication. Given how little time they
 had to plan the conference, they could have made all poster
 presentations via video-conference and call it a day. Instead, each
-poster had to record a 5-minute video summarising their
+poster had to record a 5-minute video[fn:slideslive] summarising their
 research. Alongside each presentation, there was a dedicated
 Rocket.Chat channel[fn:rocketchat] where anyone could ask a question
 to the authors, or just show their appreciation for the work. This was
@@ -62,6 +62,13 @@ All of these excellent ideas were implemented by an [[https://iclr.cc/virtual_20
 collecting all papers in a searchable, easy-to-use interface, and even
 a nice [[https://iclr.cc/virtual_2020/paper_vis.html][visualisation]] of papers as a point cloud!
 
+[fn:slideslive] The videos are streamed using [[https://library.slideslive.com/][SlidesLive]], which is a
+great solution for synchronising videos and slides. It is very
+comfortable to navigate through the slides and synchronising the video
+to the slides and vice-versa. As a result, SlidesLive also has a very
+nice library of talks, including major conferences. This is much
+better than browsing YouTube randomly.
+
 [fn:rocketchat] [[https://rocket.chat/][Rocket.Chat]] seems to be an [[https://github.com/RocketChat/Rocket.Chat][open-source]] alternative to
 Slack. Overall, the experience was great, and I appreciate the efforts
 of the organizers to use open source software instead of proprietary
@@ -135,12 +142,59 @@ very important concepts from cognitive science.
 
 TODO
 
-* Some Interesting Papers
-
-** Natural Language Processing
-
-** Reinforcement Learning
-
-** ML and Neural Network Theory
-
 * Workshops
+
+On Sunday, there were [[https://iclr.cc/virtual_2020/workshops.html][15 different workshops]]. All of them were
+recorded, and are available on the website. As always, unfortunately,
+there are too many interesting things to watch everything, but I saw
+bits and pieces of different workshops.
+
+** [[https://iclr.cc/virtual_2020/workshops_12.html][Beyond 'tabula rasa' in reinforcement learning: agents that remember, adapt, and generalize]]
+
+A lot of pretty advanced talks about RL. The general theme was
+meta-learning, aka "learning to learn". This is a very active area of
+research, which goes way beyond classical RL theory, and offer many
+interesting avenues to adjacent fields (both inside ML and outside,
+especially cognitive science). The [[http://www.betr-rl.ml/2020/abs/101/][first talk]], by Martha White, about
+inductive biases, was a very interesting and approachable introduction
+to the problems and challenges of the field. There was also a panel
+with Jürgen Schmidhuber. We hear a lot about him from the various
+controversies, but it's nice to see him talking about research and
+future developments in RL.
+
+** [[https://iclr.cc/virtual_2020/workshops_14.html][Causal Learning For Decision Making]]
+
+Ever since I read Judea Pearl's [[https://www.goodreads.com/book/show/36204378-the-book-of-why][/The Book of Why/]] on causality, I have
+been interested in how we can incorporate causality reasoning in
+machine learning. This is a complex topic, and I'm not sure yet that
+it is a complete revolution as Judea Pearl likes to portray it, but it
+nevertheless introduces a lot of new fascinating ideas. Yoshua Bengio
+gave an interesting talk[fn:bengioworkshop] (even though very similar
+to his keynote talk) on causal priors for deep learning.
+
+[fn:bengioworkshop] You can find it at 4:45:20 in the [[https://slideslive.com/38926830/workshop-on-causal-learning-for-decision-making][livestream]] of
+the workshop.
+
+** [[https://iclr.cc/virtual_2020/workshops_4.html][Bridging AI and Cognitive Science]]
+
+Cognitive science is fascinating, and I believe that collaboration
+between ML practitioners and cognitive scientists will greatly help
+advance both fields. I only watched [[https://baicsworkshop.github.io/program/baics_45.html][Leslie Kaelbling's presentation]],
+which echoes a lot of things from her talk at the main conference. It
+complements it nicely, with more focus on intelligence, especially
+/embodied/ intelligence. I think she has the rights approach to
+relationships between AI and natural science, explicitly listing the
+things from her work that would be helpful to natural scientists, and
+things she wish she knew about natural intelligences. It raises many
+fascinating questions on ourselves, what we build, and what we
+understand. I felt it was very motivational!
+
+** [[https://iclr.cc/virtual_2020/workshops_5.html][Integration of Deep Neural Models and Differential Equations]]
+
+I didn't attend this workshop, but I think I will watch the
+presentations if I can find some time. I have found the intersection
+of differential equations and ML very interesting, ever since the
+famous [[https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations][NeurIPS best paper]] on Neural ODEs. I think that such
+improvements to ML theory from other fields in mathematics would be
+extremely beneficial to a better understanding of the systems we
+build.

From 39d7826063b3c8f95816106d260573e975d95329 Mon Sep 17 00:00:00 2001
From: Dimitri Lozeve <dimitri@lozeve.com>
Date: Tue, 5 May 2020 12:15:09 +0200
Subject: [PATCH 6/6] Update wording

---
 _site/atom.xml                   | 14 +++++------
 _site/posts/iclr-2020-notes.html | 14 +++++------
 _site/rss.xml                    | 14 +++++------
 posts/iclr-2020-notes.org        | 40 ++++++++++++++++----------------
 4 files changed, 38 insertions(+), 44 deletions(-)

diff --git a/_site/atom.xml b/_site/atom.xml
index 551993a..22b280b 100644
--- a/_site/atom.xml
+++ b/_site/atom.xml
@@ -25,7 +25,7 @@
 <br />
 </span></span>.</p>
 <p>The many volunteers, the online-only nature of the event, and the low registration fees also allowed for what felt like a very diverse, inclusive event. Many graduate students and researchers from industry (like me), who do not generally have the time or the resources to travel to conferences like this, were able to attend, and make the exchanges richer.</p>
-<p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
+<p>In this post, I will try to give my impressions on the event, the speakers, and the workshops that I could attend. I will do a quick recap of the most interesting papers I saw in a future post.</p>
 <h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
 <p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
 <p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote">The videos are streamed using <a href="https://library.slideslive.com/">SlidesLive</a>, which is a great solution for synchronising videos and slides. It is very comfortable to navigate through the slides and synchronising the video to the slides and vice-versa. As a result, SlidesLive also has a very nice library of talks, including major conferences. This is much better than browsing YouTube randomly.<br />
@@ -34,9 +34,9 @@
 <br />
 </span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
 <p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
-<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
+<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even including a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
 <h1 id="speakers">Speakers</h1>
-<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
+<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw a few of them, but I expect I will be watching the others in the near future.</p>
 <h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
 <p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter. I loved the discussion on how to describe the space of distributions over domains, from the point of view of the robot factory:</p>
 <ul>
@@ -46,11 +46,9 @@
 <p>There are many ways to describe a policy (i.e. the software running in the robot’s head), and many ways to obtain them. If you are familiar with recent advances in reinforcement learning, this talk is a great occasion to take a step back, and review the relevant background ideas from engineering and control theory.</p>
 <p>Finally, the most important take-away from this talk is the importance of <em>abstractions</em>. Whatever the methods we use to program our robots, we still need a lot of human insights to give them good structural biases. There are many more insights, on the cost of experience, (hierarchical) planning, learning constraints, etc, so I strongly encourage you to watch the talk!</p>
 <h2 id="dr.-laurent-dinh-invertible-models-and-normalizing-flows">Dr. Laurent Dinh, <a href="https://iclr.cc/virtual_2020/speaker_4.html">Invertible Models and Normalizing Flows</a></h2>
-<p>This is a talk about an area of ML research I do not know very well, but very clearly presented. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
+<p>This is a very clear presentation of an area of ML research I do not know very well. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
 <h2 id="profs.-yann-lecun-and-yoshua-bengio-reflections-from-the-turing-award-winners">Profs. Yann LeCun and Yoshua Bengio, <a href="https://iclr.cc/virtual_2020/speaker_7.html">Reflections from the Turing Award Winners</a></h2>
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
-<h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
-<p>TODO</p>
 <h1 id="workshops">Workshops</h1>
 <p>On Sunday, there were <a href="https://iclr.cc/virtual_2020/workshops.html">15 different workshops</a>. All of them were recorded, and are available on the website. As always, unfortunately, there are too many interesting things to watch everything, but I saw bits and pieces of different workshops.</p>
 <h2 id="beyond-tabula-rasa-in-reinforcement-learning-agents-that-remember-adapt-and-generalize"><a href="https://iclr.cc/virtual_2020/workshops_12.html">Beyond ‘tabula rasa’ in reinforcement learning: agents that remember, adapt, and generalize</a></h2>
@@ -60,9 +58,9 @@
 <br />
 </span></span> (even though very similar to his keynote talk) on causal priors for deep learning.</p>
 <h2 id="bridging-ai-and-cognitive-science"><a href="https://iclr.cc/virtual_2020/workshops_4.html">Bridging AI and Cognitive Science</a></h2>
-<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the rights approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
+<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the right approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
 <h2 id="integration-of-deep-neural-models-and-differential-equations"><a href="https://iclr.cc/virtual_2020/workshops_5.html">Integration of Deep Neural Models and Differential Equations</a></h2>
-<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find some time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
+<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find the time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
     </section>
 </article>
 ]]></summary>
diff --git a/_site/posts/iclr-2020-notes.html b/_site/posts/iclr-2020-notes.html
index 698f893..35fd9ea 100644
--- a/_site/posts/iclr-2020-notes.html
+++ b/_site/posts/iclr-2020-notes.html
@@ -54,7 +54,7 @@
 <br />
 </span></span>.</p>
 <p>The many volunteers, the online-only nature of the event, and the low registration fees also allowed for what felt like a very diverse, inclusive event. Many graduate students and researchers from industry (like me), who do not generally have the time or the resources to travel to conferences like this, were able to attend, and make the exchanges richer.</p>
-<p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
+<p>In this post, I will try to give my impressions on the event, the speakers, and the workshops that I could attend. I will do a quick recap of the most interesting papers I saw in a future post.</p>
 <h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
 <p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
 <p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle" /><span class="sidenote">The videos are streamed using <a href="https://library.slideslive.com/">SlidesLive</a>, which is a great solution for synchronising videos and slides. It is very comfortable to navigate through the slides and synchronising the video to the slides and vice-versa. As a result, SlidesLive also has a very nice library of talks, including major conferences. This is much better than browsing YouTube randomly.<br />
@@ -63,9 +63,9 @@
 <br />
 </span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
 <p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
-<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
+<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even including a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
 <h1 id="speakers">Speakers</h1>
-<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
+<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw a few of them, but I expect I will be watching the others in the near future.</p>
 <h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
 <p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter. I loved the discussion on how to describe the space of distributions over domains, from the point of view of the robot factory:</p>
 <ul>
@@ -75,11 +75,9 @@
 <p>There are many ways to describe a policy (i.e. the software running in the robot’s head), and many ways to obtain them. If you are familiar with recent advances in reinforcement learning, this talk is a great occasion to take a step back, and review the relevant background ideas from engineering and control theory.</p>
 <p>Finally, the most important take-away from this talk is the importance of <em>abstractions</em>. Whatever the methods we use to program our robots, we still need a lot of human insights to give them good structural biases. There are many more insights, on the cost of experience, (hierarchical) planning, learning constraints, etc, so I strongly encourage you to watch the talk!</p>
 <h2 id="dr.-laurent-dinh-invertible-models-and-normalizing-flows">Dr. Laurent Dinh, <a href="https://iclr.cc/virtual_2020/speaker_4.html">Invertible Models and Normalizing Flows</a></h2>
-<p>This is a talk about an area of ML research I do not know very well, but very clearly presented. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
+<p>This is a very clear presentation of an area of ML research I do not know very well. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
 <h2 id="profs.-yann-lecun-and-yoshua-bengio-reflections-from-the-turing-award-winners">Profs. Yann LeCun and Yoshua Bengio, <a href="https://iclr.cc/virtual_2020/speaker_7.html">Reflections from the Turing Award Winners</a></h2>
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
-<h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
-<p>TODO</p>
 <h1 id="workshops">Workshops</h1>
 <p>On Sunday, there were <a href="https://iclr.cc/virtual_2020/workshops.html">15 different workshops</a>. All of them were recorded, and are available on the website. As always, unfortunately, there are too many interesting things to watch everything, but I saw bits and pieces of different workshops.</p>
 <h2 id="beyond-tabula-rasa-in-reinforcement-learning-agents-that-remember-adapt-and-generalize"><a href="https://iclr.cc/virtual_2020/workshops_12.html">Beyond ‘tabula rasa’ in reinforcement learning: agents that remember, adapt, and generalize</a></h2>
@@ -89,9 +87,9 @@
 <br />
 </span></span> (even though very similar to his keynote talk) on causal priors for deep learning.</p>
 <h2 id="bridging-ai-and-cognitive-science"><a href="https://iclr.cc/virtual_2020/workshops_4.html">Bridging AI and Cognitive Science</a></h2>
-<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the rights approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
+<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the right approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
 <h2 id="integration-of-deep-neural-models-and-differential-equations"><a href="https://iclr.cc/virtual_2020/workshops_5.html">Integration of Deep Neural Models and Differential Equations</a></h2>
-<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find some time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
+<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find the time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
     </section>
 </article>
 
diff --git a/_site/rss.xml b/_site/rss.xml
index aa71ca8..782266d 100644
--- a/_site/rss.xml
+++ b/_site/rss.xml
@@ -21,7 +21,7 @@
 <br />
 </span></span>.</p>
 <p>The many volunteers, the online-only nature of the event, and the low registration fees also allowed for what felt like a very diverse, inclusive event. Many graduate students and researchers from industry (like me), who do not generally have the time or the resources to travel to conferences like this, were able to attend, and make the exchanges richer.</p>
-<p>In this post, I will try to give my impressions on the event, and share the most interesting events and papers I saw.</p>
+<p>In this post, I will try to give my impressions on the event, the speakers, and the workshops that I could attend. I will do a quick recap of the most interesting papers I saw in a future post.</p>
 <h1 id="the-format-of-the-virtual-conference">The Format of the Virtual Conference</h1>
 <p>As a result of global travel restrictions, the conference was made fully-virtual. It was supposed to take place in Addis Ababa, Ethiopia, which is great for people who are often the target of restrictive visa policies in Northern American countries.</p>
 <p>The thing I appreciated most about the conference format was its emphasis on <em>asynchronous</em> communication. Given how little time they had to plan the conference, they could have made all poster presentations via video-conference and call it a day. Instead, each poster had to record a 5-minute video<span><label for="sn-2" class="margin-toggle sidenote-number"></label><input type="checkbox" id="sn-2" class="margin-toggle"/><span class="sidenote">The videos are streamed using <a href="https://library.slideslive.com/">SlidesLive</a>, which is a great solution for synchronising videos and slides. It is very comfortable to navigate through the slides and synchronising the video to the slides and vice-versa. As a result, SlidesLive also has a very nice library of talks, including major conferences. This is much better than browsing YouTube randomly.<br />
@@ -30,9 +30,9 @@
 <br />
 </span></span> where anyone could ask a question to the authors, or just show their appreciation for the work. This was a fantastic idea as it allowed any participant to interact with papers and authors at any time they please, which is especially important in a setting where people were spread all over the globe.</p>
 <p>There were also Zoom session where authors were available for direct, face-to-face discussions, allowing for more traditional conversations. But asking questions on the channel had also the advantage of keeping a track of all questions that were asked by other people. As such, I quickly acquired the habit of watching the video, looking at the chat to see the previous discussions (even if they happened in the middle of the night in my timezone!), and then skimming the paper or asking questions myself.</p>
-<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
+<p>All of these excellent ideas were implemented by an <a href="https://iclr.cc/virtual_2020/papers.html?filter=keywords">amazing website</a>, collecting all papers in a searchable, easy-to-use interface, and even including a nice <a href="https://iclr.cc/virtual_2020/paper_vis.html">visualisation</a> of papers as a point cloud!</p>
 <h1 id="speakers">Speakers</h1>
-<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw 4 of them, but I expect I will be watching the others in the near future.</p>
+<p>Overall, there were 8 speakers (two for each day of the main conference). They made a 40-minute presentation, and then there was a Q&amp;A both via the chat and via Zoom. I only saw a few of them, but I expect I will be watching the others in the near future.</p>
 <h2 id="prof.-leslie-kaelbling-doing-for-our-robots-what-nature-did-for-us">Prof. Leslie Kaelbling, <a href="https://iclr.cc/virtual_2020/speaker_2.html">Doing for Our Robots What Nature Did For Us</a></h2>
 <p>This talk was fascinating. It is about robotics, and especially how to design the “software” of our robots. We want to program a robot in a way that it could work the best it can over all possible domains it can encounter. I loved the discussion on how to describe the space of distributions over domains, from the point of view of the robot factory:</p>
 <ul>
@@ -42,11 +42,9 @@
 <p>There are many ways to describe a policy (i.e. the software running in the robot’s head), and many ways to obtain them. If you are familiar with recent advances in reinforcement learning, this talk is a great occasion to take a step back, and review the relevant background ideas from engineering and control theory.</p>
 <p>Finally, the most important take-away from this talk is the importance of <em>abstractions</em>. Whatever the methods we use to program our robots, we still need a lot of human insights to give them good structural biases. There are many more insights, on the cost of experience, (hierarchical) planning, learning constraints, etc, so I strongly encourage you to watch the talk!</p>
 <h2 id="dr.-laurent-dinh-invertible-models-and-normalizing-flows">Dr. Laurent Dinh, <a href="https://iclr.cc/virtual_2020/speaker_4.html">Invertible Models and Normalizing Flows</a></h2>
-<p>This is a talk about an area of ML research I do not know very well, but very clearly presented. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
+<p>This is a very clear presentation of an area of ML research I do not know very well. I really like the approach of teaching a set of methods from a “historical”, personal point of view. Laurent Dinh shows us how he arrived at this topic, what he finds interesting, in a very personal and relatable manner. This has the double advantage of introducing us to a topic that he is passionate about, while also giving us a glimpse of a researcher’s process, without hiding the momentary disillusions and disappointments, but emphasising the great achievements. Normalizing flows are also very interesting because it is grounded in strong theoretical results, that brings together a lot of different methods.</p>
 <h2 id="profs.-yann-lecun-and-yoshua-bengio-reflections-from-the-turing-award-winners">Profs. Yann LeCun and Yoshua Bengio, <a href="https://iclr.cc/virtual_2020/speaker_7.html">Reflections from the Turing Award Winners</a></h2>
 <p>This talk was very interesting, and yet felt very familiar, as if I already saw a very similar one elsewhere. Especially for Yann LeCun, who clearly reuses the same slides for many presentations at various events. They both came back to their favourite subjects: self-supervised learning for Yann LeCun, and system 1/system 2 for Yoshua Bengio. All in all, they are very good speakers, and their presentations are always insightful. Yann LeCun gives a lot of references on recent technical advances, which is great if you want to go deeper in the approaches he recommends. Yoshua Bengio is also very good at broadening the debate around deep learning, and introducing very important concepts from cognitive science.</p>
-<h2 id="prof.-michael-i.-jordan-the-decision-making-side-of-machine-learning-dynamical-statistical-and-economic-perspectives">Prof. Michael I. Jordan, <a href="https://iclr.cc/virtual_2020/speaker_8.html">The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives</a></h2>
-<p>TODO</p>
 <h1 id="workshops">Workshops</h1>
 <p>On Sunday, there were <a href="https://iclr.cc/virtual_2020/workshops.html">15 different workshops</a>. All of them were recorded, and are available on the website. As always, unfortunately, there are too many interesting things to watch everything, but I saw bits and pieces of different workshops.</p>
 <h2 id="beyond-tabula-rasa-in-reinforcement-learning-agents-that-remember-adapt-and-generalize"><a href="https://iclr.cc/virtual_2020/workshops_12.html">Beyond ‘tabula rasa’ in reinforcement learning: agents that remember, adapt, and generalize</a></h2>
@@ -56,9 +54,9 @@
 <br />
 </span></span> (even though very similar to his keynote talk) on causal priors for deep learning.</p>
 <h2 id="bridging-ai-and-cognitive-science"><a href="https://iclr.cc/virtual_2020/workshops_4.html">Bridging AI and Cognitive Science</a></h2>
-<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the rights approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
+<p>Cognitive science is fascinating, and I believe that collaboration between ML practitioners and cognitive scientists will greatly help advance both fields. I only watched <a href="https://baicsworkshop.github.io/program/baics_45.html">Leslie Kaelbling’s presentation</a>, which echoes a lot of things from her talk at the main conference. It complements it nicely, with more focus on intelligence, especially <em>embodied</em> intelligence. I think she has the right approach to relationships between AI and natural science, explicitly listing the things from her work that would be helpful to natural scientists, and things she wish she knew about natural intelligences. It raises many fascinating questions on ourselves, what we build, and what we understand. I felt it was very motivational!</p>
 <h2 id="integration-of-deep-neural-models-and-differential-equations"><a href="https://iclr.cc/virtual_2020/workshops_5.html">Integration of Deep Neural Models and Differential Equations</a></h2>
-<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find some time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
+<p>I didn’t attend this workshop, but I think I will watch the presentations if I can find the time. I have found the intersection of differential equations and ML very interesting, ever since the famous <a href="https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations">NeurIPS best paper</a> on Neural ODEs. I think that such improvements to ML theory from other fields in mathematics would be extremely beneficial to a better understanding of the systems we build.</p>
     </section>
 </article>
 ]]></description>
diff --git a/posts/iclr-2020-notes.org b/posts/iclr-2020-notes.org
index 5417b6f..c28271e 100644
--- a/posts/iclr-2020-notes.org
+++ b/posts/iclr-2020-notes.org
@@ -20,8 +20,9 @@ inclusive event. Many graduate students and researchers from industry
 travel to conferences like this, were able to attend, and make the
 exchanges richer.
 
-In this post, I will try to give my impressions on the event, and
-share the most interesting events and papers I saw.
+In this post, I will try to give my impressions on the event, the
+speakers, and the workshops that I could attend. I will do a quick
+recap of the most interesting papers I saw in a future post.
 
 [fn:volunteer] To better organize the event, and help people navigate
 the various online tools, they brought in 500(!) volunteers, waved our
@@ -60,7 +61,7 @@ skimming the paper or asking questions myself.
 
 All of these excellent ideas were implemented by an [[https://iclr.cc/virtual_2020/papers.html?filter=keywords][amazing website]],
 collecting all papers in a searchable, easy-to-use interface, and even
-a nice [[https://iclr.cc/virtual_2020/paper_vis.html][visualisation]] of papers as a point cloud!
+including a nice [[https://iclr.cc/virtual_2020/paper_vis.html][visualisation]] of papers as a point cloud!
 
 [fn:slideslive] The videos are streamed using [[https://library.slideslive.com/][SlidesLive]], which is a
 great solution for synchronising videos and slides. It is very
@@ -80,8 +81,8 @@ even avoid Zoom, because of recent privacy concerns (maybe try
 
 Overall, there were 8 speakers (two for each day of the main
 conference). They made a 40-minute presentation, and then there was a
-Q&A both via the chat and via Zoom. I only saw 4 of them, but I expect
-I will be watching the others in the near future.
+Q&A both via the chat and via Zoom. I only saw a few of them, but I
+expect I will be watching the others in the near future.
 
 ** Prof. Leslie Kaelbling, [[https://iclr.cc/virtual_2020/speaker_2.html][Doing for Our Robots What Nature Did For Us]]
 
@@ -112,12 +113,12 @@ encourage you to watch the talk!
 
 ** Dr. Laurent Dinh, [[https://iclr.cc/virtual_2020/speaker_4.html][Invertible Models and Normalizing Flows]]
 
-This is a talk about an area of ML research I do not know very well,
-but very clearly presented. I really like the approach of teaching a
-set of methods from a "historical", personal point of view. Laurent
-Dinh shows us how he arrived at this topic, what he finds interesting,
-in a very personal and relatable manner. This has the double advantage
-of introducing us to a topic that he is passionate about, while also
+This is a very clear presentation of an area of ML research I do not
+know very well. I really like the approach of teaching a set of
+methods from a "historical", personal point of view. Laurent Dinh
+shows us how he arrived at this topic, what he finds interesting, in a
+very personal and relatable manner. This has the double advantage of
+introducing us to a topic that he is passionate about, while also
 giving us a glimpse of a researcher's process, without hiding the
 momentary disillusions and disappointments, but emphasising the great
 achievements. Normalizing flows are also very interesting because it
@@ -138,9 +139,9 @@ go deeper in the approaches he recommends. Yoshua Bengio is also very
 good at broadening the debate around deep learning, and introducing
 very important concepts from cognitive science.
 
-** Prof. Michael I. Jordan, [[https://iclr.cc/virtual_2020/speaker_8.html][The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives]]
+# ** Prof. Michael I. Jordan, [[https://iclr.cc/virtual_2020/speaker_8.html][The Decision-Making Side of Machine Learning: Dynamical, Statistical and Economic Perspectives]]
 
-TODO
+# TODO
 
 * Workshops
 
@@ -182,7 +183,7 @@ between ML practitioners and cognitive scientists will greatly help
 advance both fields. I only watched [[https://baicsworkshop.github.io/program/baics_45.html][Leslie Kaelbling's presentation]],
 which echoes a lot of things from her talk at the main conference. It
 complements it nicely, with more focus on intelligence, especially
-/embodied/ intelligence. I think she has the rights approach to
+/embodied/ intelligence. I think she has the right approach to
 relationships between AI and natural science, explicitly listing the
 things from her work that would be helpful to natural scientists, and
 things she wish she knew about natural intelligences. It raises many
@@ -192,9 +193,8 @@ understand. I felt it was very motivational!
 ** [[https://iclr.cc/virtual_2020/workshops_5.html][Integration of Deep Neural Models and Differential Equations]]
 
 I didn't attend this workshop, but I think I will watch the
-presentations if I can find some time. I have found the intersection
-of differential equations and ML very interesting, ever since the
-famous [[https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations][NeurIPS best paper]] on Neural ODEs. I think that such
-improvements to ML theory from other fields in mathematics would be
-extremely beneficial to a better understanding of the systems we
-build.
+presentations if I can find the time. I have found the intersection of
+differential equations and ML very interesting, ever since the famous
+[[https://papers.nips.cc/paper/7892-neural-ordinary-differential-equations][NeurIPS best paper]] on Neural ODEs. I think that such improvements to
+ML theory from other fields in mathematics would be extremely
+beneficial to a better understanding of the systems we build.