Dissertation: PH for ML

dissertation/TDA.bib

@@ -806,4 +806,17 @@ novel application of the discriminatory power of {PIs}.},
 	date = {2017-07-17},
 	langid = {english},
 	file = {arXiv\:1706.03358 PDF:/home/dimitri/Zotero/storage/NWMEA95P/Carrière et al. - 2017 - Sliced Wasserstein Kernel for Persistence Diagrams.pdf:application/pdf;Full Text PDF:/home/dimitri/Zotero/storage/7FZZJDKP/Carrière et al. - 2017 - Sliced Wasserstein Kernel for Persistence Diagrams.pdf:application/pdf;Snapshot:/home/dimitri/Zotero/storage/VDXI2J8D/carriere17a.html:text/html}
+}
+
+@book{adler_persistent_2010,
+	title = {Persistent homology for random fields and complexes},
+	isbn = {978-0-940600-79-9},
+	url = {https://projecteuclid.org/euclid.imsc/1288099016},
+	abstract = {Project Euclid - mathematics and statistics online},
+	publisher = {Institute of Mathematical Statistics},
+	author = {Adler, Robert J. and Bobrowski, Omer and Borman, Matthew S. and Subag, Eliran and Weinberger, Shmuel},
+	urldate = {2018-07-30},
+	date = {2010},
+	doi = {10.1214/10-IMSCOLL609},
+	file = {Full Text PDF:/home/dimitri/Zotero/storage/N8UHEK9G/Adler et al. - 2010 - Persistent homology for random fields and complexe.pdf:application/pdf;Snapshot:/home/dimitri/Zotero/storage/29PXN97Q/1288099016.html:text/html}
 }

dissertation/dissertation.tex

@@ -530,12 +530,60 @@ in the evolution of the network over time.
 \label{sec:zigzag-persistence}
 
 
+
+\chapter{Persistent Homology for Machine Learning applications}%
+\label{cha:pers-homol-mach}
+
+The output of persistent homology is not directly usable by most
+statistical methods. Barcodes and persistence diagrams, being a
+multiset of points in $\overline{\mathbb{R}}^2$, are not elements of a
+metric space in which we could perform statistical computations.
+
+The distances between persistence diagrams defined
+in~\autoref{sec:topol-summ} allow us to compare different
+outputs. From a statistical perspective, it is possible to use a
+generative model of simplicial complexes, and use a distance between
+persistence diagrams to measure the similarity of our observations
+with this null model~\cite{adler_persistent_2010}. This would
+effectively define a metric space of persistence diagrams. It is even
+possible to define some statistical summaries (means, medians,
+confidence intervals) on these
+spaces~\cite{turner_frechet_2014,munch_probabilistic_2015}.
+
+The issue with this approach is that metric spaces do not offer enough
+algebraic structure to be amenable to most machine learning
+techniques. One of the most recent development in the study of
+topological summaries has been to find mappings between the space of
+persistence diagrams and Banach spaces.
+
+\section{Vectorization methods}%
+\label{sec:vect-meth}
+
+\subsection{Persistence landscapes}
+
+\subsection{Persistence images}
+
+\subsection{Tropical and arctic semirings}
+
+\section{Kernel-based methods}%
+\label{sec:kernel-based-methods}
+
+\subsection{Persistent scale-space kernel}
+
+\subsection{Persistence weighted gaussian kernel}
+
+\subsection{Sliced Wasserstein kernel}
+
+\section{Comparison}%
+\label{sec:comparison}
+
+
+
+
 \backmatter%
 
-\nocite{*}
+% \nocite{*}
-\bibliographystyle{plain}
+\printbibliography%
-\bibliography{}%
-\label{cha:bibliography}
 
 \end{document}
 

dissertation/preamble.tex

@@ -33,7 +33,7 @@
 \usepackage{tikz}
 \usetikzlibrary{patterns,backgrounds,positioning}
 
-\usepackage{biblatex}
+\usepackage[style=numeric-comp]{biblatex}
 \bibliography{TDA,temporalgraphs}
 
 \usepackage{pdfpages}