Skip to content

Commit 8932e2c

Browse files
committed
updates ppl and pubs
1 parent 1ed5745 commit 8932e2c

File tree

6 files changed

+113
-48
lines changed

6 files changed

+113
-48
lines changed

_data/people.yml

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -15,13 +15,15 @@ faculty:
1515
scholar: "https://scholar.google.fr/citations?user=muGuSyoAAAAJ"
1616
twitter: "https://x.com/GBR_data"
1717
github: "https://github.com/bricaud"
18+
linkedin: https://www.linkedin.com/in/benjaminricaud/
1819

1920

2021
phd:
2122
- name: "Michele Guerra"
2223
image: "/figs/people/mg.png"
2324
bio: "Michele is a PhD student researching randomized and trainable models for time series and graphs. The focus of is research is on uncertainty quantification and interpretability."
2425
scholar: https://scholar.google.ca/citations?user=HzMTPo0AAAAJ&hl=en&oi=ao
26+
linkedin: https://www.linkedin.com/in/michele-guerra-34a273182/
2527

2628
- name: "Jonas Berg Hansen"
2729
bio: "Jonas is a PhD student researching pooling in graph neural networks."

_data/publications.yml

Lines changed: 15 additions & 15 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,17 @@
1+
- title: "MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks"
2+
authors: "Carlo Abate, Filippo Maria Bianchi"
3+
figure: "figs/publications/maxcutpool.png"
4+
abstract: "We propose a novel approach to compute the MAXCUT in attributed graphs, i.e., graphs with features associated with nodes and edges. Our approach is robust to the underlying graph topology and is fully differentiable, making it possible to find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement a hierarchical graph pooling layer for Graph Neural Networks, which is sparse, differentiable, and particularly suitable for downstream tasks on heterophilic graphs."
5+
github: "https://github.com/NGMLGroup/MaxCutPool"
6+
arxiv: "https://arxiv.org/abs/2409.05100"
7+
bibtex: |
8+
@article{abate2024maxcutpool,
9+
title={MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks},
10+
author={Abate, Carlo and Bianchi, Filippo Maria},
11+
journal={arXiv preprint arXiv:2409.05100},
12+
year={2024}
13+
}
14+
115
- title: "Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling"
216
authors: "Ivan Marisca, Cesare Alippi, Filippo Maria Bianchi"
317
venue: "ICML 2024"
@@ -15,18 +29,4 @@
1529
volume = {235},
1630
series = {Proceedings of Machine Learning Research},
1731
publisher = {PMLR}
18-
}
19-
20-
- title: "MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks"
21-
authors: "Carlo Abate, Filippo Maria Bianchi"
22-
figure: "figs/publications/maxcutpool.png"
23-
abstract: "We propose a novel approach to compute the MAXCUT in attributed graphs, i.e., graphs with features associated with nodes and edges. Our approach is robust to the underlying graph topology and is fully differentiable, making it possible to find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement a hierarchical graph pooling layer for Graph Neural Networks, which is sparse, differentiable, and particularly suitable for downstream tasks on heterophilic graphs."
24-
github: "https://github.com/NGMLGroup/MaxCutPool"
25-
arxiv: "https://arxiv.org/abs/2409.05100"
26-
bibtex: |
27-
@article{abate2024maxcutpool,
28-
title={MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks},
29-
author={Abate, Carlo and Bianchi, Filippo Maria},
30-
journal={arXiv preprint arXiv:2409.05100},
31-
year={2024}
32-
}
32+
}

_site/feed.xml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,2 +1,2 @@
1-
<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.3.4">Jekyll</generator><link href="http://localhost:4000/feed.xml" rel="self" type="application/atom+xml" /><link href="http://localhost:4000/" rel="alternate" type="text/html" /><updated>2024-10-14T12:39:42+02:00</updated><id>http://localhost:4000/feed.xml</id><title type="html">Northernmost GraphML Group</title><subtitle>The Northermost GraphML group in the world, based in Tromsø, Norway.
1+
<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="4.3.4">Jekyll</generator><link href="http://localhost:4000/feed.xml" rel="self" type="application/atom+xml" /><link href="http://localhost:4000/" rel="alternate" type="text/html" /><updated>2024-10-14T13:44:30+02:00</updated><id>http://localhost:4000/feed.xml</id><title type="html">Northernmost GraphML Group</title><subtitle>The Northermost GraphML group in the world, based in Tromsø, Norway.
22
</subtitle></feed>

_site/people.html

Lines changed: 16 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -50,6 +50,7 @@
5050
<div class="people-page">
5151
<h1>Meet our team!</h1>
5252

53+
<!-- Faculty Members -->
5354
<section class="faculty">
5455
<h2>Faculty Members</h2>
5556

@@ -112,6 +113,10 @@ <h3>Benjamin Ricaud</h3>
112113
</a>
113114

114115

116+
<a href="https://www.linkedin.com/in/benjaminricaud/" target="_blank" title="LinkedIn">
117+
<i class="fab fa-linkedin"></i>
118+
</a>
119+
115120

116121
<a href="https://github.com/bricaud" target="_blank" title="GitHub">
117122
<i class="fab fa-github"></i>
@@ -123,6 +128,12 @@ <h3>Benjamin Ricaud</h3>
123128

124129
</section>
125130

131+
<!-- Postdoc Researchers -->
132+
<section class="postdoc">
133+
<h2>Researchers and postdocs</h2>
134+
135+
</section>
136+
126137
<!-- PhD Students -->
127138
<section class="phd">
128139
<h2>PhD students</h2>
@@ -142,6 +153,10 @@ <h3>Michele Guerra</h3>
142153

143154

144155

156+
<a href="https://www.linkedin.com/in/michele-guerra-34a273182/" target="_blank" title="LinkedIn">
157+
<i class="fab fa-linkedin"></i>
158+
</a>
159+
145160
</div>
146161
</div>
147162
</div>
@@ -163,6 +178,7 @@ <h3>Jonas Berg Hansen</h3>
163178

164179
</section>
165180

181+
<!-- Collaborators -->
166182
<section class="collaborators"></section>
167183
<h2>Collaborators and visitors</h2>
168184

_site/publications.html

Lines changed: 31 additions & 32 deletions
Original file line numberDiff line numberDiff line change
@@ -55,39 +55,35 @@ <h1>Publications</h1>
5555
<div class="publication">
5656

5757
<!-- Title -->
58-
<h3 class="pub-title"><strong>Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling</strong></h3>
58+
<h3 class="pub-title"><strong>MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks</strong></h3>
5959

6060
<!-- Authors -->
6161
<p class="authors">
62-
<em style="color: gray;">Ivan Marisca, Cesare Alippi, Filippo Maria Bianchi</em>
62+
<em style="color: gray;">Carlo Abate, Filippo Maria Bianchi</em>
6363
</p>
6464

6565
<!-- Venue (optional) -->
6666

67-
<p class="venue">
68-
<em>ICML 2024</em>
69-
</p>
70-
7167

7268
<!-- Figure -->
7369

7470
<div class="pub-figure">
75-
<img src="figs/publications/hdtts.png" alt="Figure for Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling">
71+
<img src="figs/publications/maxcutpool.png" alt="Figure for MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks">
7672
</div>
7773

7874

7975
<!-- Abstract -->
80-
<p>Spatiotemporal graph neural networks achieve striking results by representing the relationships across time series as a graph. Nonetheless, most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden spatiotemporal dynamics when part of the data is missing. In this work, we tackle this problem through hierarchical spatiotemporal downsampling. The input time series are progressively coarsened over time and space, obtaining a pool of representations that capture heterogeneous temporal and spatial dynamics. Conditioned on observations and missing data patterns, such representations are combined by an interpretable attention mechanism to generate the forecasts.</p>
76+
<p>We propose a novel approach to compute the MAXCUT in attributed graphs, i.e., graphs with features associated with nodes and edges. Our approach is robust to the underlying graph topology and is fully differentiable, making it possible to find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement a hierarchical graph pooling layer for Graph Neural Networks, which is sparse, differentiable, and particularly suitable for downstream tasks on heterophilic graphs.</p>
8177

8278
<!-- Links (GitHub, Arxiv, Cite) -->
8379
<div class="pub-links">
8480

85-
<a href="https://github.com/marshka/hdtts" target="_blank" class="btn">
81+
<a href="https://github.com/NGMLGroup/MaxCutPool" target="_blank" class="btn">
8682
<i class="fab fa-github"></i> GitHub
8783
</a>
8884

8985

90-
<a href="https://arxiv.org/abs/2402.10634" target="_blank" class="btn">
86+
<a href="https://arxiv.org/abs/2409.05100" target="_blank" class="btn">
9187
<i class="fas fa-file-alt"></i> ArXiv
9288
</a>
9389

@@ -99,15 +95,11 @@ <h3 class="pub-title"><strong>Graph-based Forecasting with Missing Data through
9995

10096
<!-- BibTeX Entry (Hidden by Default) -->
10197
<div id="bibtex-entry-1" class="bibtex-entry" style="display: none;">
102-
<pre>@inproceedings{marisca2024graph,
103-
title = {Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling},
104-
author = {Marisca, Ivan and Alippi, Cesare and Bianchi, Filippo Maria},
105-
booktitle = {Proceedings of the 41st International Conference on Machine Learning},
106-
pages = {34846--34865},
107-
year = {2024},
108-
volume = {235},
109-
series = {Proceedings of Machine Learning Research},
110-
publisher = {PMLR}
98+
<pre>@article{abate2024maxcutpool,
99+
title={MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks},
100+
author={Abate, Carlo and Bianchi, Filippo Maria},
101+
journal={arXiv preprint arXiv:2409.05100},
102+
year={2024}
111103
}
112104
</pre>
113105
</div>
@@ -117,35 +109,39 @@ <h3 class="pub-title"><strong>Graph-based Forecasting with Missing Data through
117109
<div class="publication">
118110

119111
<!-- Title -->
120-
<h3 class="pub-title"><strong>MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks</strong></h3>
112+
<h3 class="pub-title"><strong>Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling</strong></h3>
121113

122114
<!-- Authors -->
123115
<p class="authors">
124-
<em style="color: gray;">Carlo Abate, Filippo Maria Bianchi</em>
116+
<em style="color: gray;">Ivan Marisca, Cesare Alippi, Filippo Maria Bianchi</em>
125117
</p>
126118

127119
<!-- Venue (optional) -->
128120

121+
<p class="venue">
122+
<em>ICML 2024</em>
123+
</p>
124+
129125

130126
<!-- Figure -->
131127

132128
<div class="pub-figure">
133-
<img src="figs/publications/maxcutpool.png" alt="Figure for MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks">
129+
<img src="figs/publications/hdtts.png" alt="Figure for Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling">
134130
</div>
135131

136132

137133
<!-- Abstract -->
138-
<p>We propose a novel approach to compute the MAXCUT in attributed graphs, i.e., graphs with features associated with nodes and edges. Our approach is robust to the underlying graph topology and is fully differentiable, making it possible to find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement a hierarchical graph pooling layer for Graph Neural Networks, which is sparse, differentiable, and particularly suitable for downstream tasks on heterophilic graphs.</p>
134+
<p>Spatiotemporal graph neural networks achieve striking results by representing the relationships across time series as a graph. Nonetheless, most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden spatiotemporal dynamics when part of the data is missing. In this work, we tackle this problem through hierarchical spatiotemporal downsampling. The input time series are progressively coarsened over time and space, obtaining a pool of representations that capture heterogeneous temporal and spatial dynamics. Conditioned on observations and missing data patterns, such representations are combined by an interpretable attention mechanism to generate the forecasts.</p>
139135

140136
<!-- Links (GitHub, Arxiv, Cite) -->
141137
<div class="pub-links">
142138

143-
<a href="https://github.com/NGMLGroup/MaxCutPool" target="_blank" class="btn">
139+
<a href="https://github.com/marshka/hdtts" target="_blank" class="btn">
144140
<i class="fab fa-github"></i> GitHub
145141
</a>
146142

147143

148-
<a href="https://arxiv.org/abs/2409.05100" target="_blank" class="btn">
144+
<a href="https://arxiv.org/abs/2402.10634" target="_blank" class="btn">
149145
<i class="fas fa-file-alt"></i> ArXiv
150146
</a>
151147

@@ -157,13 +153,16 @@ <h3 class="pub-title"><strong>MaxCutPool: differentiable feature-aware Maxcut fo
157153

158154
<!-- BibTeX Entry (Hidden by Default) -->
159155
<div id="bibtex-entry-2" class="bibtex-entry" style="display: none;">
160-
<pre>@article{abate2024maxcutpool,
161-
title={MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks},
162-
author={Abate, Carlo and Bianchi, Filippo Maria},
163-
journal={arXiv preprint arXiv:2409.05100},
164-
year={2024}
165-
}
166-
</pre>
156+
<pre>@inproceedings{marisca2024graph,
157+
title = {Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling},
158+
author = {Marisca, Ivan and Alippi, Cesare and Bianchi, Filippo Maria},
159+
booktitle = {Proceedings of the 41st International Conference on Machine Learning},
160+
pages = {34846--34865},
161+
year = {2024},
162+
volume = {235},
163+
series = {Proceedings of Machine Learning Research},
164+
publisher = {PMLR}
165+
}</pre>
167166
</div>
168167

169168
</div>

people.html

Lines changed: 48 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,7 @@
66
<div class="people-page">
77
<h1>Meet our team!</h1>
88

9+
<!-- Faculty Members -->
910
<section class="faculty">
1011
<h2>Faculty Members</h2>
1112
{% for person in site.data.people.faculty %}
@@ -51,6 +52,52 @@ <h3>{{ person.name }}</h3>
5152
{% endfor %}
5253
</section>
5354

55+
<!-- Postdoc Researchers -->
56+
<section class="postdoc">
57+
<h2>Researchers and postdocs</h2>
58+
{% for person in site.data.people.postdoc %}
59+
<div class="person">
60+
<img src="{% if person.image %}{{ person.image }}{% else %}{{'/figs/people/default.png'}}{% endif %}" alt="{{ person.name }}'s profile picture" class="profile-pic" />
61+
<div class="info">
62+
<h3>{{ person.name }}</h3>
63+
<p>{{ person.bio }}</p>
64+
<div class="links">
65+
{% if person.webpage %}
66+
<a href="{{ person.webpage }}" target="_blank" title="Personal Webpage">
67+
<i class="fas fa-globe"></i>
68+
</a>
69+
{% endif %}
70+
{% if person.email %}
71+
<a href="mailto:{{ person.email }}" title="Email">
72+
<i class="fas fa-envelope"></i>
73+
</a>
74+
{% endif %}
75+
{% if person.scholar %}
76+
<a href="{{ person.scholar }}" target="_blank" title="Google Scholar">
77+
<i class="fas fa-graduation-cap"></i>
78+
</a>
79+
{% endif %}
80+
{% if person.twitter %}
81+
<a href="{{ person.twitter }}" target="_blank" title="Twitter">
82+
<i class="fa-brands fa-x-twitter"></i>
83+
</a>
84+
{% endif %}
85+
{% if person.linkedin %}
86+
<a href="{{ person.linkedin }}" target="_blank" title="LinkedIn">
87+
<i class="fab fa-linkedin"></i>
88+
</a>
89+
{% endif %}
90+
{% if person.github %}
91+
<a href="{{ person.github }}" target="_blank" title="GitHub">
92+
<i class="fab fa-github"></i>
93+
</a>
94+
{% endif %}
95+
</div>
96+
</div>
97+
</div>
98+
{% endfor %}
99+
</section>
100+
54101
<!-- PhD Students -->
55102
<section class="phd">
56103
<h2>PhD students</h2>
@@ -92,6 +139,7 @@ <h3>{{ person.name }}</h3>
92139
{% endfor %}
93140
</section>
94141

142+
<!-- Collaborators -->
95143
<section class="collaborators"></section>
96144
<h2>Collaborators and visitors</h2>
97145
{% for person in site.data.people.collaborator %}

0 commit comments

Comments
 (0)