-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex3.html
286 lines (274 loc) · 18.4 KB
/
index3.html
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
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
<!DOCTYPE html>
<html lang="en">
<head>
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-127803291-1"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-127803291-1');
</script>
<title>News – Bayesian Methods Research Group</title>
<meta property="twitter:site" content="@bayesgroup" />
<meta property="twitter:card" content="summary" />
<meta property="og:title" content="News – Bayesian Methods Research Group" />
<meta property="og:description" content="from Bremen, Germany" />
<meta property="og:image" content="http://bayesgroup.org/theme/images/bayesgroup_opengraph.png" />
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<!--[if lte IE 8]><script src="/theme/js/ie/html5shiv.js"></script><![endif]-->
<link rel="stylesheet" href="/theme/css/main.css" />
<!--[if lte IE 8]><link rel="stylesheet" href="/theme/css/ie8.css" /><![endif]-->
</head>
<body>
<!-- Header -->
<section id="header">
<header>
<span class="image"><a href="/"><img src="/theme/images/bayesgroup.png" alt="" /></a></span>
<h1 id="logo">Bayesian Methods Research Group</h1>
<p>from Bremen, Germany</p>
</header>
<nav id="nav">
<ul>
<li><a href="/about/">About</a></li>
<li><a href="/people">People</a></li>
<li><a href="/publications">Publications</a></li>
<li><a href="/teaching">Teaching</a></li>
<li><a href="/admission/">Admission</a></li>
<li><a href="/alumni/">Alumni</a></li>
</ul>
</nav>
<footer>
<ul class="icons">
<li><a href="https://twitter.com/bayesgroup" class="icon fa-twitter"><span class="label">Twitter</span></a></li>
<li><a href="https://www.youtube.com/@bayesgroup" class="icon fa-youtube"><span class="label">Youtube</span></a></li>
<li><a href="mailto:[email protected]" class="icon fa-envelope"><span class="label">Email</span></a></li>
</ul>
</footer>
</section>
<!-- Wrapper -->
<div id="wrapper">
<!-- Group Photo -->
<section id="photo">
<div class="container">
<span class="image fit"><img src="/theme/images/banner.jpg" alt="" /></span>
</div>
</section>
<!-- Main -->
<div id="main">
<section id="content">
<div class="container">
<header class="major">
<h2>News</h2>
</header>
<ol id="news-list">
<li>
<article class="entry">
<header>
<h3 class="entry-title">AAAI 2020</h3>
<span class="entry-date">
<span class="entry-date-day">11</span>
<span class="entry-date-month">Dec</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p>Two papers have been accepted to the Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI 2020):</p>
<ul>
<li><a href="/publications/2020-structured-sparsification-of-gated-recurrent-neural-networks/">Structured Sparsification of Gated Recurrent Neural Networks</a> by Ekaterina Lobacheva, Nadezhda Chirkova, Aleksandr Markovich, and Dmitry Vetrov</li>
<li><a href="/publications/2020-low-variance-black-box-gradient-estimates-for-the-plackett-luce-distribution/">Low-variance Black-box Gradient Estimates for the Plackett-Luce Distribution</a> by Artyom Gadetsky, Kirill Struminsky, Dmitry Vetrov in collaboration with Christopher Robinson and Novi Quadrianto. </li>
</ul> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">NeurIPS 2019 Workshops</h3>
<span class="entry-date">
<span class="entry-date-day">25</span>
<span class="entry-date-month">Oct</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p>We've got several papers accepted to NeurIPS workshops:</p>
<ol>
<li>"Pitfalls of In-Domain Uncertainty Estimation and Ensembling in Deep Learning" by Arsenii Ashukha, Alexander Lyzhov, Dmitry Molchanov and Dmitry Vetrov has been accepted to the <a href="http://bayesiandeeplearning.org">Bayesian Deep Learning Workshop</a>.</li>
<li>"Low-variance Gradient Estimates for the Plackett-Luce Distribution" by Artyom Gadetsky, Kirill Struminsky, Novi Quadrianto and Dmitry Vetrov in collaboration with Christopher Robinson has been accepted to the Bayesian Deep Learning Workshop.</li>
<li>"Unsupervised Domain Adaptation with Shared Latent Dynamics for Reinforcement Learning" by Evgenii Nikishin, Arsenii Ashukha and Dmitry Vetrov has also been accepted to the Bayesian Deep Learning Workshop.</li>
<li>"Structured Sparsification of Gated Recurrent Neural Networks" by Ekaterina Lobacheva, Nadezhda Chirkova, Alexander Markovich and Dmitry Vetrov has been accepted to the workshop on <a href="https://context-composition.github.io">Context and Compositionality in Biological and Artificial Neural Systems</a>.</li>
<li>Finally, Max Kochurov contributed to the "<a href="https://openreview.net/forum?id=rkgzj5Za8H">PyMC4: Exploiting Coroutines for Implementing a Probabilistic Programming Framework</a>" paper accepted to the workshop on <a href="https://program-transformations.github.io">Program Transformations</a>.</li>
</ol> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">NeurIPS 2019</h3>
<span class="entry-date">
<span class="entry-date-day">12</span>
<span class="entry-date-month">Sep</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p>This year we've doubled our presence at NeurIPS with four papers accepted:</p>
<ol>
<li><a href="/publications/2019-importance-weighted-hierarchical-variational-inference/">Importance Weighted Hierarchical Variational Inference</a> by Artem Sobolev and Dmitry Vetrov.</li>
<li><a href="/publications/2019-the-implicit-metropolis-hastings-algorithm/">The Implicit Metropolis-Hastings Algorithm</a> by Kirill Neklyudov and Dmitry Vetrov in collaboration with Evgenii Egorov.</li>
<li><a href="/publications/2019-a-simple-baseline-for-bayesian-uncertainty-in-deep-learning/">A Simple Baseline for Bayesian Uncertainty in Deep Learning</a> by Timur Garipov and Dmitry Vetrov in collaboration with Wesley Maddox, Pavel Izmailov and Andrew Gordon Wilson.</li>
<li><a href="/publications/2019-a-prior-of-a-googol-gaussians:-a-tensor-ring-induced-prior-for-generative-models/">A Prior of a Googol Gaussians: a Tensor Ring Induced Prior for Generative Models</a> by Maxim Kuznetsov, Daniil Polykovskiy and Dmitry Vetrov in collaboration with Alexander Zhebrak.</li>
</ol>
<p>Good academic service is not only about producing novel research, but also about providing critical assessment of other's work.
We're proud that Kirill Struminsky, Ekaterina Lobacheva, Dmitry Molchanov, Arsenii Ashukha, Dmitry Vetrov and Dmitry Kropotov were recognized as top 50% reviewers.</p> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">A paper published in Nature Biotechnology</h3>
<span class="entry-date">
<span class="entry-date-day">02</span>
<span class="entry-date-month">Sep</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p><a href="https://insilico.com/">Insilico Medicine</a> published an article in Nature Biotechnology coauthored by our members Maxim Kuznetsov and Daniil Polykovskiy. The <a href="/publications/2019-deep-learning-enables-rapid-identification-of-potent-ddr1-kinase-inhibitors/">paper</a> describes a timed challenge, where the new machine learning system called Generative Tensorial Reinforcement Learning (GENTRL) designed six novel inhibitors of DDR1, a kinase target implicated in fibrosis and other diseases, in 21 days. Four compounds were active in biochemical assays, and two were validated in cell-based assays. One lead candidate was tested and demonstrated favorable pharmacokinetics in mice.</p> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">Deep|Bayes 2019 is over</h3>
<span class="entry-date">
<span class="entry-date-day">29</span>
<span class="entry-date-month">Aug</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p>It's this time of a year: once again students from all over the world gathered in Moscow to participate in the Deep|Bayes 2019 – a summer school on Bayesian Deep Learning. Just like the last year, the school featured both lectures and practical assignments. We've been also fortunate to have a couple of invited speakers: <a href="http://users.sussex.ac.uk/~nq28/">Novi Quadrianto</a> from University of Sussex and Higher School of Economics, <a href="http://www.eurecom.fr/~filippon/">Maurizio Filippone</a> from EURECOM, <a href="https://franrruiz.github.io/">Francisco Jesus Rodriguez Ruiz</a> from Columbia University and University of Cambridge, <a href="http://mi.eng.cam.ac.uk/~am969/">Andrey Malinin</a> from University of Cambridge and Sergey Bartunov from DeepMind.</p>
<p>For an official press-release see the <a href="https://cs.hse.ru/en/big-data/bayeslab/news/303676800.html">HSE website</a>. The slides, videos and practicals are available at <a href="http://deepbayes.ru/">deepbayes.ru</a></p> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">Call for Postdoc on Deep RL</h3>
<span class="entry-date">
<span class="entry-date-day">02</span>
<span class="entry-date-month">Apr</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p>We are looking for a postdoc to join our group! Please see details <a href="/news/call-for-postdoc-on-deep-rl/">here</a>.</p>
</div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">Applications to Deep|Bayes 2019 are now open</h3>
<span class="entry-date">
<span class="entry-date-day">16</span>
<span class="entry-date-month">Feb</span>
<span class="entry-date-year">2019</span>
</span>
</header>
<div class="entry-content"> <p>One again, we're organizing an international summer school on Bayesian Deep Learning to be held in Moscow, August 20–25. Head over to <a href="http://deepbayes.ru/">deepbayes.ru</a> to view last year's videos, practical assignments and apply to this year's run.</p> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">ICLR 2019</h3>
<span class="entry-date">
<span class="entry-date-day">21</span>
<span class="entry-date-month">Dec</span>
<span class="entry-date-year">2018</span>
</span>
</header>
<div class="entry-content"> <p>We got 3 papers accepted to ICLR 2019:</p>
<ul>
<li><a href="/publications/2019-variational-autoencoder-with-arbitrary-conditioning/">Variational Autoencoder with Arbitrary Conditioning</a> by Oleg Ivanov, Michael Figurnov and Dmitry Vetrov;</li>
<li><a href="/publications/2019-variance-networks-when-expectation-does-not-meet-your-expectations/">Variance Networks: When Expectation Does Not Meet Your Expectations</a> by Kirill Neklyudov, Dmitry Molchanov, Arsenii Ashukha, Dmitry Vetrov;</li>
<li><a href="/publications/2019-the-deep-weight-prior/">The Deep Weight Prior</a> by Andrei Atanov, Arsenii Ashukha, Kirill Struminsky, Dmitriy Vetrov in collaboration with Max Welling.</li>
</ul> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">NeurIPS 2018 Results</h3>
<span class="entry-date">
<span class="entry-date-day">09</span>
<span class="entry-date-month">Dec</span>
<span class="entry-date-year">2018</span>
</span>
</header>
<div class="entry-content"> <p>This year's NeurIPS conference turned out to be a very fruitful one! We've had</p>
<ul>
<li>Two papers accepted, one of them being a spotlight<ol>
<li><a href="https://papers.nips.cc/paper/7347-quantifying-learning-guarantees-for-convex-but-inconsistent-surrogates">Quantifying Learning Guarantees for Convex but Inconsistent Surrogates</a> by Kirill Struminsky and Anton Osokin in collaboration with Simon Lacoste-Julien</li>
<li><a href="https://papers.nips.cc/paper/8095-loss-surfaces-mode-connectivity-and-fast-ensembling-of-dnns">Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNs</a> by Timur Garipov, Dmitrii Podoprikhin and Dmitry Vetrov in collaboration with Pavel Izmailov and Andrew Gordon Wilson</li>
</ol>
</li>
<li>An invited talk by Dmitry Vetrov at the <a href="http://bayesiandeeplearning.org/">Bayesian Deep Learning</a> workshop</li>
<li>Three papers accepted to the aforementioned workshop<ol>
<li><a href="http://bayesiandeeplearning.org/2018/papers/36.pdf">Importance Weighted Hierarchical Variational Inference</a> by Artem Sobolev and Dmitry Vetrov</li>
<li><a href="http://bayesiandeeplearning.org/2018/papers/35.pdf">Variational Dropout via Empirical Bayes</a> by Valery Kharitonov, Dmitry Molchanov and Dmitry Vetrov</li>
<li><a href="http://bayesiandeeplearning.org/2018/papers/55.pdf">Subset-Conditioned Generation Using Variational Autoencoder With A Learnable Tensor-Train Induced Prior</a> by Maksim Kuznetsov, Daniil Polykovskiy and Dmitry Vetrov in collaboration with Alexander Zhebrak</li>
</ol>
</li>
<li>A paper accepted to the <a href="https://sites.google.com/site/rlponips2018">Reinforcement Learning under Partial Observability</a> workshop (contributed talk)<ol>
<li><a href="https://www.ias.informatik.tu-darmstadt.de/uploads/Team/JoniPajarinen/RLPO2018_paper_28.pdf">Joint Belief Tracking and Reward Optimization through Approximate Inference</a> by Pavel Shvechikov, Alexander Grishin, Arseny Kuznetsov, Alexander Fritzler and Dmitry Vetrov</li>
</ol>
</li>
<li>A paper accepted to the <a href="https://openreview.net/group?id=NIPS.cc/2018/Workshop/CDNNRIA#accepted-papers">Compact Deep Neural Network Representation with Industrial Applications</a> workshop<ol>
<li><a href="https://openreview.net/forum?id=ByMQgZHYoX">Bayesian Sparsification of Gated Recurrent Neural Networks</a> by Ekaterina Lobacheva, Nadezhda Chirkova and Dmitry Vetrov</li>
</ol>
</li>
</ul> </div><!-- /.entry-content -->
</article>
</li>
<li>
<article class="entry">
<header>
<h3 class="entry-title">Three papers accepted to ACML 2018</h3>
<span class="entry-date">
<span class="entry-date-day">10</span>
<span class="entry-date-month">Nov</span>
<span class="entry-date-year">2018</span>
</span>
</header>
<div class="entry-content"> <p>We have 3 papers accepted to ACML 2018:
<a href="http://proceedings.mlr.press/v95/polykovskiy18a.html">Concorde: Morphological Agreement in Conversational Models</a> by Daniil Polykovskiy in collaboration with Dmitry Soloviev and Sergey Nikolenko,
<a href="http://proceedings.mlr.press/v95/kemaev18a.html">ReSet: Learning Recurrent Dynamic Routing in ResNet-like Neural Networks</a> by Iurii Kemaev, Daniil Polykovskiy and Dmitry Vetrov, and
<a href="http://proceedings.mlr.press/v95/kuzminykh18a.html">Extracting Invariant Features From Images Using An Equivariant Autoencoder</a> by Daniil Polykovskiy in collaboration with Denis Kuzminykh, Alexander Zhebrak.</p> </div><!-- /.entry-content -->
</article>
</li>
</ol><!-- /#posts-list -->
<p class="paginator">
<a href="/index2.html" class="paginator-new">← Newer</a>
Page 3 out of 4
<a href="/index4.html" class="paginator-old">Older →</a>
</p>
</div>
</section>
</div>
<!-- Footer -->
<section id="footer">
<div class="container">
<ul class="copyright">
<li>Proudly powered by <a href="http://getpelican.com/">Pelican</a>, which takes great advantage of <a href="http://python.org">Python</a>.</li><li>Design: <a href="http://html5up.net">HTML5 UP</a></li>
</ul>
</div>
</section>
</div>
<!-- Scripts -->
<script src="/theme/js/jquery.min.js"></script>
<script src="/theme/js/jquery.scrollzer.min.js"></script>
<script src="/theme/js/jquery.scrolly.min.js"></script>
<script src="/theme/js/skel.min.js"></script>
<script src="/theme/js/util.js"></script>
<!--[if lte IE 8]><script src="theme/js/ie/respond.min.js"></script><![endif]-->
<script src="/theme/js/main.js"></script>
</body>
</html>