-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.xml
168 lines (133 loc) · 9.88 KB
/
index.xml
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
<?xml version="1.0" encoding="utf-8" standalone="yes" ?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<title>Jeremias Knoblauch on Jeremias Knoblauch</title>
<link>https://jeremiasknoblauch.github.io/</link>
<description>Recent content in Jeremias Knoblauch on Jeremias Knoblauch</description>
<generator>Hugo -- gohugo.io</generator>
<language>en-us</language>
<copyright>&copy; 2018</copyright>
<lastBuildDate>Fri, 16 Feb 2018 00:00:00 -0500</lastBuildDate>
<atom:link href="/" rel="self" type="application/rss+xml" />
<item>
<title>New horizons for Bayesian inference</title>
<link>https://jeremiasknoblauch.github.io/research_projects/new-horizons-for-bayesian-inference/</link>
<pubDate>Mon, 08 Apr 2019 00:00:00 -0400</pubDate>
<guid>https://jeremiasknoblauch.github.io/research_projects/new-horizons-for-bayesian-inference/</guid>
<description><p>Inspired by Zellner (1988) and Bissiri, Holmes &amp; Walker (2016), this project aims at extending the flexibility and customizeability of Bayesian inference. The first stepping stone on this path is Generalized Variational Inference (GVI) which I coauthored together with Jack Jewson &amp; Theodoros Damoulas. GVI is based on a generalized representation of Bayesian inference that I believe can inspire further methodological advances. Key for this project is going to be a solid information-geometric understanding of Bayesian inference and notions of discrepancy in the space of distributions. As we have shown, this approach is remarkably powerful (see e.g., Knoblauch et al. (2018) for an application to Bayesian On-line Changepoint Detection or Knoblauch (2019) for an application to Deep Gaussian Processes).</p>
</description>
</item>
<item>
<title>Generalized Variational Inference</title>
<link>https://jeremiasknoblauch.github.io/publication/gvi/</link>
<pubDate>Mon, 08 Apr 2019 00:00:00 +0000</pubDate>
<guid>https://jeremiasknoblauch.github.io/publication/gvi/</guid>
<description></description>
</item>
<item>
<title>Robust Deep Gaussian Processes</title>
<link>https://jeremiasknoblauch.github.io/publication/gvi-dgp/</link>
<pubDate>Mon, 08 Apr 2019 00:00:00 +0000</pubDate>
<guid>https://jeremiasknoblauch.github.io/publication/gvi-dgp/</guid>
<description></description>
</item>
<item>
<title>Bayesian On-line Changepoint Detection 2.0</title>
<link>https://jeremiasknoblauch.github.io/research_projects/advancing-bayesian-on-line-changepoint-detection/</link>
<pubDate>Mon, 07 May 2018 00:00:00 -0400</pubDate>
<guid>https://jeremiasknoblauch.github.io/research_projects/advancing-bayesian-on-line-changepoint-detection/</guid>
<description><p><strong>Bayesian On-line Changepoint Detection (BOCPD)</strong> is a discrete-time inference framework introduced in the statistics and machine learning community independently by <a href="http://eprints.lancs.ac.uk/745/1/online_chpt4.pdf" target="_blank">Fearnhead &amp; Liu (2007)</a> and <a href="https://arxiv.org/abs/0710.3742" target="_blank">Adams &amp; MacKay (2007)</a>. Taken together, both papers have generated in excess of 500 citations and inspired more research in this area. The method is popular because it is efficient and runs in constant time per observation processed. Me and my collaborators are working on extending the inference paradigm in several ways:</p>
<ul class="task-list">
<li><label><input type="checkbox" checked disabled class="task-list-item"> Unifiying Fearnhead &amp; Liu (2007) and Adams &amp; MacKay (2007)¹</label></li>
<li><label><input type="checkbox" checked disabled class="task-list-item"> Multivariate analysis¹</label></li>
<li><label><input type="checkbox" checked disabled class="task-list-item"> Robust analysis²</label></li>
<li><label><input type="checkbox" disabled class="task-list-item"> Continuous-time models</label></li>
<li><label><input type="checkbox" disabled class="task-list-item"> Point processes</label></li>
</ul>
<p>The software written as part of this ongoing project received the <a href="https://github.com/alan-turing-institute/ReproducibleResearchResources" target="_blank">Alan Turing Institute&rsquo;s Reproducible Research award</a>, and will be linked here via my github repo once it failsafe to use.</p>
<p>¹Jeremias Knoblauch, Theodoros Damoulas. <a href="https://jeremiasknoblauch.github.io/publication/stbocpdms/" target="_blank">Spatio-temporal Bayesian On-line Changepoint Detection</a>, <em>International Conference on Machine Learning</em> (2018).</p>
<p>²
Jeremias Knoblauch, Jack Jewson, Theodoros Damoulas. <a href="https://jeremiasknoblauch.github.io/publication/robustbocpd/" target="_blank">Doubly Robust Bayesian Inference for Non-Stationary Streaming Data with β-Divergences</a>, <em>arXiv:1806.02261</em> (2018).</p>
</description>
</item>
<item>
<title>Post Selection Inference in Weakly Stationary Time Series</title>
<link>https://jeremiasknoblauch.github.io/research_projects/robust-post-model-selection-inference/</link>
<pubDate>Mon, 07 May 2018 00:00:00 -0400</pubDate>
<guid>https://jeremiasknoblauch.github.io/research_projects/robust-post-model-selection-inference/</guid>
<description><p>This work is not yet published, but Stephan and I are hoping to resume our work on it soon. If you have a strong interest in this subject, do get in touch with me and I will happily share my thesis work with you.</p>
</description>
</item>
<item>
<title>Doubly Robust Bayesian Inference for Non-Stationary Streaming Data with β-Divergences</title>
<link>https://jeremiasknoblauch.github.io/publication/robustbocpd/</link>
<pubDate>Mon, 07 May 2018 00:00:00 +0000</pubDate>
<guid>https://jeremiasknoblauch.github.io/publication/robustbocpd/</guid>
<description></description>
</item>
<item>
<title>Spatio-Temporal Bayesian On-line Changepoint Detection with Model Selection</title>
<link>https://jeremiasknoblauch.github.io/publication/stbocpdms/</link>
<pubDate>Fri, 09 Feb 2018 00:00:00 +0000</pubDate>
<guid>https://jeremiasknoblauch.github.io/publication/stbocpdms/</guid>
<description></description>
</item>
<item>
<title>Bayesian Analysis for Non-Stationary Streaming Data</title>
<link>https://jeremiasknoblauch.github.io/talk/edinburgh_cdt/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/edinburgh_cdt/</guid>
<description><p>Embed your slides or video here using <a href="https://sourcethemes.com/academic/post/writing-markdown-latex/" target="_blank">shortcodes</a>. Further details can easily be added using <em>Markdown</em> and $\rm \LaTeX$ math code.</p>
</description>
</item>
<item>
<title>Bayesian On-line Changepoint Detection and Model Selection in high-dimensional data</title>
<link>https://jeremiasknoblauch.github.io/talk/edinburgh_icms/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/edinburgh_icms/</guid>
<description></description>
</item>
<item>
<title>Doubly Robust Bayesian Inference for Non-Stationary Streaming Data with β-Divergences</title>
<link>https://jeremiasknoblauch.github.io/talk/fbtechtalk/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/fbtechtalk/</guid>
<description></description>
</item>
<item>
<title>Doubly Robust Bayesian Inference for Non-Stationary Streaming Data with β-Divergences</title>
<link>https://jeremiasknoblauch.github.io/talk/nips2018/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/nips2018/</guid>
<description></description>
</item>
<item>
<title>Doubly Robust Bayesian Inference for Non-Stationary Streaming Data with β-Divergences</title>
<link>https://jeremiasknoblauch.github.io/talk/oxwaspsymposium/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/oxwaspsymposium/</guid>
<description></description>
</item>
<item>
<title>Generalized Variational Inference</title>
<link>https://jeremiasknoblauch.github.io/talk/columbia/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/columbia/</guid>
<description></description>
</item>
<item>
<title>Generalized Variational Inference</title>
<link>https://jeremiasknoblauch.github.io/talk/cornell/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/cornell/</guid>
<description></description>
</item>
<item>
<title>Generalized Variational Inference</title>
<link>https://jeremiasknoblauch.github.io/talk/nyu/</link>
<pubDate>Sun, 01 Jan 2017 00:00:00 -0500</pubDate>
<guid>https://jeremiasknoblauch.github.io/talk/nyu/</guid>
<description></description>
</item>
</channel>
</rss>