Ron Levie (@levie_ron) header
Ron Levie (@levie_ron) profile picture

Ron Levie

@levie_ron

Senior Lecturer (Assistant Professor) at the Faculty of Mathematics, Technion – Israel Institute of Technology @TechnionLive

 Haifa, Israel
sites.google.com/view/ronlevie
Joined June 2021
  • Tweets23
  • Following72
  • Followers191
1Media
Photo by Ron Levie (@levie_ron)
Spectral and message-passing GNNs are often studied by different communities, with little interaction. In arxiv.org/abs/2602.10031, we argue that the two architecture types have distinct strengths and should be studied in a unified way.
Image from Christopher Morris (@chrsmrrs)
1
4
8
836
Want to know about the current understanding of the generalization abilities of GNNs? Please have a look at our survey paper arxiv.org/abs/2503.15650. Joint work with Antonis Vasileiou, Stefanie Jegelka, and @levie_ron.
Image from Christopher Morris (@chrsmrrs)
1
36
2
155
11,654
Check out our #NeurIPS2024 paper: Equivariant Machine Learning on Graphs with Nonlinear Spectral Filters arxiv.org/abs/2406.01249 With Ronen Talmon and @levie_ron, we consider graph functional shifts as a symmetry group and propose NLSFs that are equivariant to these shifts
Image from Ya-Wei Eileen Lin (@YaWeiEileenLin)
4
21
1,111
If you're a Game of Thrones ⚔️ or Graph Learning fan, I'll present "Learning on Large Graphs using Intersecting Communities" @NeurIPS 📌East Exhibit Hall A-C #3001, Session 3 ⏲️Thu 12 Dec 11-14 Reach out to chat about Geometric DL! Thanks @ismaililkanc @mmbronstein @levie_ron
Image from Ben Finkelshtein (@benfinkelshtein)
10
32
2,798
Learning on Large Graphs using Intersecting Communities is accepted to @NeurIPSConf #neurips2024! An efficient algorithm with linear memory/time following our constructive Weak Graph Regularity Lemma Thanks to @ismaililkanc @mmbronstein @levie_ron Paper:arxiv.org/abs/2405.20724
1
12
40
5,377
Check out my #NeurIPS paper “A Graphon-Signal Analysis of GNNs.” I introduce a metric on the space of graphs with node features, which reveals that the input space of GNNs has Euclidean-like properties. This allows us to derive nice analysis like generalization theorems.
Image from Ron Levie (@levie_ron)
2
12
82
10,165
Link to the paper: arxiv.org/abs/2305.15987 Link to the poster, for NeurIPS participants: neurips.cc/virtual/2023/post…
A graphon-signal analysis of graph neural networks

A graphon-signal analysis of graph neural networks

We present an approach for analyzing message passing graph neural networks (MPNNs) based on an extension of graphon analysis to a so called graphon-signal analysis. A MPNN is a function that takes...

arxiv.org
3
481
GNNs typically exploit permutation symmetry. Yet for learning tasks on a fixed graph, we show that enforcing active/approximate symmetries improves generalization. Check out our work "Approximately Equivariant Graph Networks" #NeurIPS23 (joint work w/ @levie_ron @SoledadVillar5 )
Image from Teresa Huang (@TeresaNHuang)
1
13
62
8,194
1-WL enhanced our understanding of the ultimate limitations of GNNs. However, it doesn't reveal similarities between graphs in the feature space. Check out our new NeurIPS paper, where we develop a continuous variant to better understand the topology of GNNs' feature space.
Image from Christopher Morris (@chrsmrrs)
2
29
2
137
19,311
3/3 accepted papers at #NeurIPS2023 ! In the papers, with @chrsmrrs, @SoledadVillar5, @TeresaNHuang and Jan Böker, we analyze GNNs by considering meaningful notions of graph metrics using graphon theory, which give structure to the domain of definition of GNNs. By that: 1/3
3
5
49
6,707
We prove new generalization and universal approximation theorems, alignment of GNNs with graph metrics inspired by 1WL, and novel GNN architectures which are equivariant to approximate symmetries. 2/3
1
1
10
1,040
arxiv.org/abs/2306.03698 arxiv.org/abs/2308.10436 arxiv.org/abs/2305.15987 3/3
Fine-grained Expressivity of Graph Neural Networks

Fine-grained Expressivity of Graph Neural Networks

Numerous recent works have analyzed the expressive power of message-passing graph neural networks (MPNNs), primarily utilizing combinatorial techniques such as the $1$-dimensional Weisfeiler-Leman...

arxiv.org
2
14
728
Generalization bounds in #DeepLearning are typically: capacity / dataset size For message passing NNs, @SohirMaskey, Y. Lee, @GittaKutyniok and I proved the bound: capacity / (dataset size * size of graphs) facilitating generalization on large graphs. arxiv.org/abs/2202.00645
2
2
28
* Here, we deal with inductive learning on a dataset of graphs. The term (dataset size) is the number of graphs in the training set. * The (size of graphs) term is the average number of nodes in graphs in the dataset, to some positive power.
1
2
* (capacity) is some complexity measure of the hypothesis class of message passing networks
2
I am thrilled to announce that I will be starting a tenure-track faculty position at the Faculty of Mathematics, @TechnionLive. Thank you to everyone who supported me and to all of my amazing collaborators!
2
1
21
Load more