-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathpapers.html
More file actions
253 lines (225 loc) · 17.4 KB
/
papers.html
File metadata and controls
253 lines (225 loc) · 17.4 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
<!DOCTYPE HTML>
<!--
Spatial by TEMPLATED
templated.co @templatedco
Released for free under the Creative Commons Attribution 3.0 license (templated.co/license)
-->
<html>
<head>
<title>Xenia Miscouridou - Papers</title>
<meta http-equiv="content-type" content="text/html; charset=utf-8" />
<meta name="description" content="Papers" />
<meta name="keywords" content="papers,Emile Mathieu,Mathieu emile,emilemathieu" />
<link rel="shortcut icon" href="Images/graph.png" type="image/vnd.microsoft.icon" />
<!--[if lte IE 8]><script src="js/html5shiv.js"></script><![endif]-->
<!--<script src="js/jquery.min.js"></script>
<script src="js/skel.min.js"></script>
<script src="js/skel-layers.min.js"></script>
<script src="js/init.js"></script>
-->
<script src="https://cdnjs.cloudflare.com/ajax/libs/jquery/3.1.1/jquery.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/skel/2.2.1/skel.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/skel-layers/2.0.2/skel-layers.min.js"></script>
<script src="js/init.js"></script>
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','https://www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-90612705-1', 'auto');
ga('send', 'pageview');
</script>
<noscript>
<!-- <link rel="stylesheet" href="css/skel.css" /> -->
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/skel/2.2.1/skel.css" />
<link rel="stylesheet" href="css/style.css" />
<link rel="stylesheet" href="css/style-xlarge.css" />
</noscript>
</head>
<body class="landing">
<!-- Header -->
<header id="header" class="alt">
<nav id="share" style="left: 1.25em;">
<ul class="icons">
<li style="margin-left:1em"><a href="https://github.com/misxenia" target="_blank" class="icon fa-github"><span class="label">Github</span></a></li>
<li style="margin-left:1em"><a href="https://scholar.google.co.uk/citations?user=eUsxsWwAAAAJ&hl=en" target="_blank" class="icon fa-google-plus-square"><span class="label">Linkedin</span></a></li>
<li style="margin-left:1em"><a href="https://www.linkedin.com/in/xenia-miscouridou-69388582" target="_blank" class="icon fa-linkedin-square"><span class="label">Linkedin</span></a></li>
</ul>
</nav>
<!--<h1><strong><a href="index.html">Emile Mathieu</a></strong></h1>-->
<nav id="nav">
<ul>
<li><a href="index.html">Home</a></li>
<li><a href="education.html">education</a><li>
<li><strong><a href="papers.html">Papers</a></strong/></li>
<li><a href="software.html">software</a><li>
<li><a href="contact.html">Contact</a>></li>
</ul>
</nav>
</header>
<!-- Banner -->
<section id="banner">
<h2>Xenia Miscouridou</h2>
</section>
<!-- misc -->
<section id="papers" class="wrapper">
<div class="container" style="max-width: 1000px; line-height: 140%;">
<header class="major special">
<h2>Papers</h2>
<!-- <p>Mainly class projects</p> -->
</header>
<p id="RMHMC">
<a href="PDFs/RMHMC_presentation.pdf" target="_blank">
<!-- <span class="image left">
<img src="Images/graph.png" title="Exchangeable Random Measures for " alt="Hamiltonian Monte Carlo" style="max-width: 120px">
</span> -->
</a>
<h2> Papers and Preprints </h2>
<blockquote>Exchangeable Random Measures for Sparse Graphs with Overlapping Communities </i> , Adrien Todeschini, <b>Xenia Miscouridou,</b> Francois Caron <a href="https://arxiv.org/abs/1602.02114" target="_blank"> arXiv:1602.02114</a> ,2017<br>
</blockquote>
We propose a novel statistical model for sparse networks with overlapping community structure. The model is based on representing the graph as an exchangeable point process, and naturally generalizes existing probabilistic models with overlapping block-structure to the sparse regime. Our construction builds on vectors of completely random measures, and has interpretable parameters, each node being assigned a vector representing its level of affiliation to some latent communities. We develop methods for simulating this class of random graphs, as well as to perform posterior inference. We show that the proposed approach can recover interpretable structure from two real-world networks and can handle graphs with thousands of nodes and tens of thousands of edges.
</br> <!-- [<a href="PDFs/RMHMC_presentation.pdf" target="_blank">Presentation</a>] [<a href="https://github.com/emilemathieu/RiemannHamiltonianMonteCarlo" target="_blank">Sources</a>] -->
</p>
<!--
<p id="policySearchReview">
<a href="PDFs/policysearchreport.pdf" target="_blank">
<span class="image left">
<img src="Images/gps.jpg" title="Guided Policy Search" alt="Guided Policy Search" style="max-width: 120px">
</span>
</a>
<h4>
Policy Search: A review
<a href="works.html#policySearchReview"><span class="image right"><img src="Images/link-to-here.png" title="Link to this paper" alt="Link to this paper" style="max-width: 15px"></span></a>
</h4>
<blockquote>
This is the result of a project part of the <a href="http://researchers.lille.inria.fr/~lazaric/Webpage/MVA-RL_Course16.html" target="_blank">Reinforcement Learning course</a> taught by <a href="http://researchers.lille.inria.fr/~lazaric/" target="_blank">A.Lazaric</a>. This project's goal was to write a review on Policy Search, a reinforcement learning's sub-field which proceeds by directly learning a parametrized policy without estimated any value function.
</blockquote>
Charles Reizine, Emile Mathieu, 2016<br>
[<a href="PDFs/policysearchreport.pdf" target="_blank">Report</a>] [<a href="PDFs/policysearchpresentation.pdf" target="_blank">Presentation</a>]
</p>
<p id="factorialHiddenMarkovmodel">
<a href="PDFs/fhmmposter.pdf" target="_blank">
<span class="image left">
<img src="Images/fhmm2.png" title="Factorial hidden Markov models" alt="Factorial hidden Markov models" style="max-width: 120px">
</span>
</a>
<h4>
Factorial Hidden Markov Models
<a href="works.html#factorialHiddenMarkovmodel"><span class="image right"><img src="Images/link-to-here.png" title="Link to this paper" alt="Link to this paper" style="max-width: 15px"></span></a>
</h4>
<blockquote>
I studied in depth the <a href="http://www.ee.columbia.edu/~sfchang/course/svia-F03/papers/factorial-HMM-97.pdf" target="_blank">factorial hidden Markov models article</a> wrote by Zoubin Ghahramani and Michael I. Jordan as part of the <a href="http://imagine.enpc.fr/~obozinsg/teaching/mva_gm/fall2016/" target="_blank">Probabilistic Graphical Model course</a> taught by <a href="http://imagine.enpc.fr/~obozinsg/" target="_blank">Guillaume Obozinski</a> and <a href="http://www.di.ens.fr/~fbach" target="_blank">Francis Bach</a>. Factorial hidden Markov models are an extension of classical hidden Markov models with a distributed representation of their state's space.
</blockquote>
Emile Mathieu, 2016<br>
[<a href="PDFs/fhmmreport.pdf" target="_blank">Report</a>] [<a href="PDFs/fhmmposter.pdf" target="_blank">Poster</a>]
</p>
<p id="gpbandits">
<a href="PDFs/gpbanditsreport.pdf" target="_blank">
<span class="image left">
<img src="Images/gp.png" title="Gaussian process bandits" alt="Gaussian process bandits" style="max-width: 120px">
</span>
</a>
<h4>
Gaussian Process Bandits with Thompson Sampling
<a href="works.html#gpbandits"><span class="image right"><img src="Images/link-to-here.png" title="Link to this paper" alt="Link to this paper" style="max-width: 15px"></span></a>
</h4>
<blockquote>
As part of the <a href="http://researchers.lille.inria.fr/~valko/hp/mva-ml-graphs.php" target="_blank">Graphs in ML course</a> taught by <a href="http://researchers.lille.inria.fr/~valko/" target="_blank">Michal Valko</a>, I proposed and implemented a Thompson Sampling algorithm for the Gaussian Process bandit setting. This setting is described in <a href="http://www-stat.wharton.upenn.edu/~skakade/papers/ml/bandit_GP_icml.pdf" target="_blank">Gaussian Process Optimization in the Bandit Setting: No Regret and Experimental Design</a>.
</blockquote>
Emile Mathieu, 2016<br>
[<a href="PDFs/gpbanditsreport.pdf" target="_blank">Report</a>] [<a href="PDFs/gpbanditspresentation.pdf" target="_blank">Presentation</a>] [<a href="https://github.com/emilemathieu/Project_GP-bandits" target="_blank">Sources</a>]
</p>
<p id="stagescientifique">
<a href="PDFs/Rapport_Scientifique_light.pdf" target="_blank">
<span class="image left">
<img src="Images/bss.png" title="Bicycle sharing system" alt="Bicycle sharing system" style="max-width: 120px">
</span>
</a>
<h4>
Research Internship: Urban mobility data analysis
<a href="works.html#stagescientifique"><span class="image right"><img src="Images/link-to-here.png" title="Link to this paper" alt="Link to this paper" style="max-width: 15px"></span></a>
</h4>
<blockquote>
During an internship at <a href="http://www.ifsttar.fr/" target="_blank">Ifsttar</a>, I applied probabilistic models such as LDA, along with web visualizations, to transportation’s
data in order to enhance the understanding of commuters’ behavior.
</blockquote>
Emile Mathieu, 2014<br>
[<a href="PDFs/Rapport_Scientifique_light.pdf" target="_blank">Report</a>]
</p> -->
<p id="talks and posters">
<!-- <a href="PDFs/malapreport.pdf" target="_blank"> -->
<!-- <span class="image left">
<img src="Images/graph.png" title="Mixture of experts" alt="Mixture of experts" style="max-width: 120px">
</span> -->
</a>
<h2> Talks and Posters </h2>
</p>
<p>
<p id="paris">
<!-- <a href="PDFs/malapreport.pdf" target="_blank"> -->
<blockquote>Sparse Graphs with Overlapping Communities </blockquote>
<h5> Contributed Talk at the <a href="https://www.ceremade.dauphine.fr/~salomond/BNP11/" target="_blank"> 11th Conference of Bayesian NonParametrics</a> </br> Paris, June 2017
</h5>
<!-- <a href="works.html#learningfromcrowds"><span class="image right"><img src="Images/link-to-here.png" title="Link to this paper" alt="Link to this paper" style="max-width: 15px"></span></a> -->
<!-- [<a href="PDFs/malapreport.pdf" target="_blank">Report</a>] [<a href="https://github.com/emilemathieu/MALAP_Projet" target="_blank">Sources</a>] -->
<p id="milan">
<!-- <a href="PDFs/malapreport.pdf" target="_blank"> -->
<blockquote>Sparse Modular Graphs with Overlapping Communities </blockquote>
<h5> Poster at the <a href="https://www.unibocconi.eu/wps/wcm/connect/Site/BISP10/Home" target="_blank"> 10th International Conference on Bayesian Inference in Stochastic Processes (BISP)</a> </br> Milan, June 2017
</h5>
<!-- [<a href="PDFs/malapreport.pdf" target="_blank">Report</a>] [<a href="https://github.com/emilemathieu/MALAP_Projet" target="_blank">Sources</a>] -->
<!--
<p id="AMAZONposter">
<a href="PDFs/malapreport.pdf" target="_blank">
<span class="image left">
<img src="Images/graph.png" title="Mixture of experts" alt="Mixture of experts" style="max-width: 120px">
</span>
</a>
<blockquote>Sparse Modular Graphs with Overlapping Communities </blockquote>
<h5> Poster at the Machine Learning Workshop - Amazon)</a> </br> Berlin, March 2017
</h5> -->
</div>
</section>
<!-- contact -->
<!-- <section id="contact" class="wrapper style1 special">
<div class="container">
<header class="special">
<h2>Contact me</h2>
</header>
<div class="feature-grid">
<div class="feature">
<div class="image small rounded" ><img src="Images/email2.png" alt=""/></div>
<div class="content small">
<header>
<h4>Email</h4>
</header>
xenia.miscouridou-at-spc.ox.ac.uk
</div>
</div>
<div class="feature">
<div class="image rounded small"><img src="Images/post2.png" alt="" /></div>
<div class="content small">
<header>
<h4>Post</h4>
</header>
Department of Statistics<br>
University of Oxford<br>
24-29 St Giles<br>
Oxford OX1 3LB, UK
</div>
</div>
</div>
</div>
</section> -->
<!-- Footer -->
<footer id="footer">
<div class="container">
<ul class="copyright">
<li>Copyright © Xenia Miscouridou</li>
<!-- <li>Adapted Design: Copyright © <a href="http://http://mlg.eng.cam.ac.uk/yarin/" target="_blank">Yarin Gal</a></li> -->
<li>Original Template Design: <a href="http://templated.co" target="_blank">Spatial from TEMPLATED</a></li>
</ul>
</div>
</footer>
</body>
</html>