-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
266 lines (237 loc) · 16 KB
/
index.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
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Wolfgang Stammer - Symbolic Concepts, Explanations and Interactions ... and stuff</title>
<link rel="stylesheet" href="styles.css">
<link href="https://fonts.googleapis.com/css2?family=Architects+Daughter&family=Roboto:wght@300&display=swap" rel="stylesheet">
</head>
<body>
<header>
<nav>
<ul>
<li><a href="#about" class="active">About</a></li>
<li><a href="#publications">Publications</a></li>
<li><a href="#music">Music</a></li>
</ul>
<div class="social-icons">
<a href="https://scholar.google.com/citations?user=66-aU5AAAAAJ&hl" target="blank">
<img src="logos/icons8-google-scholar.svg" alt="Google Scholar" style="height: 24px;">
</a>
<a href="https://twitter.com/WolfStammer" target="blank">
<img src="logos/icons8-twitter-48.png" alt="Twitter" style="height: 24px;">
</a>
<a href="https://de.linkedin.com/in/wolfgang-stammer" target="blank">
<img src="logos/icons8-linkedin-48.png" alt="LinkedIn" style="height: 24px;">
</a>
<a href="https://github.com/wolfstam" target="blank">
<img src="logos/icons8-github-24.png" alt="GitHub" style="height: 24px;">
</a>
</div>
</nav>
</header>
<main>
<div class="intro">
<img src="images/wstammer3.jpeg" alt="Your Photo">
<div class="intro-text">
<h1>Wolfgang Stammer</h1>
<h2>Symbolic Concepts, Explanations and Interactions</h2>
<p>PhD Student</p>
<p>Technical University Darmstadt</p>
<p>AI & ML lab</p>
<p>wolfgang dot stammer at cs dot tu minus darmstadt dot de</p>
</div>
</div>
<section id="about">
<p>My research revolves around the question: "How can we create <span style="color: #759fE5;">performative</span> AI models that help users <span style="color:#a5d781;">understand</span> their decision-making processes and <span style="color:#6d57d1;">interact</span> with their <span style="color:#DF9D9B;">internal representations</span>, enabling revisions of <span style="color:#f0d677;">shortcut</span> behaviors?"</p>
<p>I believe that for AI systems to explain their decisions effectively, they must communicate with human stakeholders using verifiable, concept-level statements. Importantly, this communication should not be one-sided; like human conversations, it should involve active discussion and interaction.</p>
<p>As part of this, I'm also interested in how cognitive systems—whether biological or artificial—can learn abstract concepts without strong supervision. How can they bind information to a specific representation? What kind of representation is it?</p>
<p>Cool events I helped with:</p>
<ul>
<li>"Semantic, Symbolic and Interpretable Machine Learning" ELLIS Workshop at Elise Wrap Up Conference 2024</li>
<li>"Interactive Machine Learning" Workshop at AAAI 2022</li>
<li>"Explanations in Interactive Machine Learning" Tutorial at AAAI 2022</li>
<li>"Perspectives on Learning" Doctoral Symposium on Cognitive Science of <i>German Society for Cognitive Science</i>, 2022</li>
</ul>
<p>Apart from research I am very passionate about writing and playing music (see below :)).</p>
</section>
<section id="publications">
<h2>Selected Publications</h2>
<p>Show me in <a href="https://scholar.google.com/citations?user=66-aU5AAAAAJ&hl" class="scholar-link">Google Scholar</a></p>
<p></p>
<!-- Displayed Publications -->
<!-- <div id="selected-publications"> -->
<div class="publication-item">
<img src="papers/stammer2024neurips.png" alt="Publication Image">
<div class="publication-text">
<h3>Neural Concept Binder</h3>
<p><u>Wolfgang Stammer</u>, Antonia Wüst, David Steinmann, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2406.09949" class="publication-link">[NeurIPS 2024]</a>
<a href="https://github.com/ml-research/NeuralConceptBinder" class="github-link">[GitHub]</a>
</div>
</div>
<!-- <div id="hidden-publications" style="display: none;"></div> -->
<div class="publication-item">
<img src="papers/delfosse2024interpretable.png" alt="Publication Image">
<div class="publication-text">
<h3>Interpretable Concept Bottlenecks to Align Reinforcement Learning Agents</h3>
<p>Quentin Delfosse, Sebastian Sztwiertnia, Mark Rothermel, <u>Wolfgang Stammer</u>, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2401.05821" class="publication-link">[NeurIPS 2024]</a>
<a href="https://github.com/k4ntz/SCoBots" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/wuest_pix2code.png" alt="Publication Image">
<div class="publication-text">
<h3>Pix2code: Learning to compose neural visual concepts as programs</h3>
<p>Antonia Wüst, <u>Wolfgang Stammer</u>, Quentin Delfosse, Devendra Singh Dhami, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2402.08280" class="publication-link">[UAI 2024] (oral)</a>
<a href="https://github.com/ml-research/pix2code" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/stammer2024tmlr.png" alt="Publication Image">
<div class="publication-text">
<h3>Learning by Self-Explaining</h3>
<p><u>Wolfgang Stammer</u>, Felix Friedrich, David Steinmann, Hikaru Shindo, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2309.08395" class="publication-link">[TMLR 2024]</a>
<a href="https://github.com/ml-research/learning-by-self-explaining" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/helff2023vlol.png" alt="Publication Image">
<div class="publication-text">
<h3>V-LoL: A Diagnostic Dataset for Visual Logical Learning</h3>
<p>Lukas Helff, <u>Wolfgang Stammer</u>, Hikaru Shindo, Devendra Singh Dhami, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2306.07743" class="publication-link">[DMLR 2024]</a>
<a href="https://sites.google.com/view/v-lol" class="project-page">[Project Page]</a>
<a href="https://huggingface.co/datasets/AIML-TUDA/v-lol-trains" class="huggingface-link">[HuggingFace]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/steinemann2024intervene.png" alt="Publication Image">
<div class="publication-text">
<h3>Learning to Intervene on Concept Bottlenecks</h3>
<p>David Steinmann, <u>Wolfgang Stammer</u>, Felix Friedrich, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2308.13453" class="publication-link">[ICML 2024]</a>
<a href="https://github.com/ml-research/CB2M" class="github-link">[GitHub]</a>
</div>
</div>
<!-- </div> -->
<div class="publication-item">
<img src="papers/busch_whereisthetruth.png" alt="Publication Image">
<div class="publication-text">
<h3>Where is the Truth? The Risk of Getting Confounded in a Continual World</h3>
<p>Florian Peter Busch, Roshni Kamath, Rupert Mitchell, <u>Wolfgang Stammer</u>, Kristian Kersting, Martin Mundt</p>
<a href="https://arxiv.org/pdf/2402.06434" class="publication-link">[arxiv 2024]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/teso2023frontiers.png" alt="Publication Image">
<div class="publication-text">
<h3>Leveraging explanations in interactive machine learning: An overview</h3>
<p>Stefano Teso, Öznur Alkan, <u>Wolfgang Stammer</u>, Elizabeth Daly</p>
<a href="https://www.frontiersin.org/articles/10.3389/frai.2023.1066049/full" class="publication-link">[Frontiers in AI 2023]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/friedrich2023xiltypology.png" alt="Publication Image">
<div class="publication-text">
<h3>A typology for exploring the mitigation of shortcut behaviour</h3>
<p>Felix Friedrich, <u>Wolfgang Stammer</u>, Patrick Schramowski, Kristian Kersting</p>
<a href="https://www.nature.com/articles/s42256-023-00612-w" class="publication-link">[Nature Machine Intelligence 2023]</a>
<a href="https://github.com/ml-research/A-Typology-for-Exploring-the-Mitigation-of-Shortcut-Behavior" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/delfosse2023ecml.png" alt="Publication Image">
<div class="publication-text">
<h3>Boosting Object Representation Learning via Motion and Object Continuity</h3>
<p>Quentin Delfosse, <u>Wolfgang Stammer</u>, Thomas Rothenbacher, Dwarak Vittal, Kristian Kersting</p>
<a href="https://arxiv.org/pdf/2211.09771" class="publication-link">[ECML-PKDD 2023]</a>
<a href="https://github.com/k4ntz/MOC" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/friedrich2022RiT.png" alt="Publication Image">
<div class="publication-text">
<h3>Revision Transformers: Instructing Language Models to Change Their Values</h3>
<p>Felix Friedrich, <u>Wolfgang Stammer</u>, Patrick Schramowski, Kristian Kersting</p>
<a href="https://ebooks.iospress.nl/pdf/doi/10.3233/FAIA230341" class="publication-link">[ECAI 2023]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/stammer2022cvpr.png" alt="Publication Image">
<div class="publication-text">
<h3>Interactive Disentanglement: Learning Concepts by Interacting with their Prototype Representations</h3>
<p><u>Wolfgang Stammer</u>, Marius Memmel, Patrick Schramowski, Kristian Kersting</p>
<a href="https://openaccess.thecvf.com/content/CVPR2022/papers/Stammer_Interactive_Disentanglement_Learning_Concepts_by_Interacting_With_Their_Prototype_Representations_CVPR_2022_paper.pdf" class="publication-link">[CVPR 2022]</a>
<a href="https://github.com/ml-research/XIConceptLearning" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/skryagin2022kr_slash.png" alt="Publication Image">
<div class="publication-text">
<h3>Neural-Probabilistic Answer Set Programming</h3>
<p>Arseny Skryagin, <u>Wolfgang Stammer</u>, Daniel Ochs, Devendra Singh Dhami, Kristian Kersting</p>
<a href="https://proceedings.kr.org/2022/48/kr2022-0048-skryagin-et-al.pdf" class="publication-link">[KR 2022]</a>
<a href="https://github.com/ml-research/SLASH" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/shao2021aaai.png" alt="Publication Image">
<div class="publication-text">
<h3>Right for better reasons: Training differentiable models by constraining their influence functions</h3>
<p>Xiaoting Shao, Arseny Skryagin, <u>Wolfgang Stammer</u>, Patrick Schramowski, Kristian Kersting</p>
<a href="https://ojs.aaai.org/index.php/AAAI/article/view/17148/16955" class="publication-link">[AAAI 2021]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/stammer2021cvpr.png" alt="Publication Image">
<div class="publication-text">
<h3>Right for the Right Concept: Revising Neuro-Symbolic Concepts by Interacting With Their Explanations</h3>
<p><u>Wolfgang Stammer</u>, Patrick Schramowski, Kristian Kersting</p>
<a href="http://openaccess.thecvf.com/content/CVPR2021/papers/Stammer_Right_for_the_Right_Concept_Revising_Neuro-Symbolic_Concepts_by_Interacting_CVPR_2021_paper.pdf" class="publication-link">[CVPR 2021]</a>
<a href="https://github.com/ml-research/NeSyXIL" class="github-link">[GitHub]</a>
</div>
</div>
<div class="publication-item">
<img src="papers/schramowski2020arxiv_plantxml.png" alt="Publication Image">
<div class="publication-text">
<h3>Making deep neural networks right for the right scientific reasons by interacting with their explanations</h3>
<p>Patrick Schramowski, <u>Wolfgang Stammer</u>, Stefano Teso, Anna Brugger, Franziska Herbert, Xiaoting Shao, Hans-Georg Luigs, Anne-Katrin Mahlein, Kristian Kersting</p>
<a href="https://www.nature.com/articles/s42256-020-0212-3" class="publication-link">[Nature Machine Intelligence 2020]</a>
<a href="https://github.com/ml-research/XIL" class="github-link">[GitHub]</a>
</div>
</div>
<!-- </div> -->
<!-- Show More Button
<p id="toggle-publications" onclick="togglePublications()" style="cursor: pointer; color: blue; text-decoration: underline;">Show all</p> -->
</section>
<section id="music">
<div class="music">
<h2>Music</h2>
<img src="images/Spiderwebs.jpeg" alt="Spiderwebs & Foam">
<p>In addition to my work in AI and machine learning, I am also passionate about music. I play and write lyrics and music in the semi-professional band <a href="https://spiderwebsandfoam.com" target="_blank">Spiderwebs & Foam</a>. Our band blends various genres, from Rock, Jazz, Electronic, Folk creating a unique sound that resonates with our thoughts :D. We have performed at several venues and continue to create and share our music with the world.</p>
</div>
</section>
</main>
<footer>
<p>© 2024 Wolfgang Stammer. All rights reserved.</p>
</footer>
<!-- <script>
function togglePublications() {
var hiddenPublications = document.getElementById("hidden-publications");
var toggleButton = document.getElementById("toggle-publications");
if (hiddenPublications.style.display === "none") {
hiddenPublications.style.display = "block";
toggleButton.textContent = "Show less";
} else {
hiddenPublications.style.display = "none";
toggleButton.textContent = "Show all";
}
}
</script> -->
</body>
</html>