Cooperative Graph Neural Networks

Ben Finkelshtein, Xingyue Huang, Michael M. Bronstein, Ismail Ilkan Ceylan
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:13633-13659, 2024.

Abstract

Graph neural networks are popular architectures for graph machine learning, based on iterative computation of node representations of an input graph through a series of invariant transformations. A large class of graph neural networks follow a standard message-passing paradigm: at every layer, each node state is updated based on an aggregate of messages from its neighborhood. In this work, we propose a novel framework for training graph neural networks, where every node is viewed as a player that can choose to either listen, broadcast, listen and broadcast, or to isolate. The standard message propagation scheme can then be viewed as a special case of this framework where every node listens and broadcasts to all neighbors. Our approach offers a more flexible and dynamic message-passing paradigm, where each node can determine its own strategy based on their state, effectively exploring the graph topology while learning. We provide a theoretical analysis of the new message-passing scheme which is further supported by an extensive empirical analysis on a synthetic and real-world datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-finkelshtein24a, title = {Cooperative Graph Neural Networks}, author = {Finkelshtein, Ben and Huang, Xingyue and Bronstein, Michael M. and Ceylan, Ismail Ilkan}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {13633--13659}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/finkelshtein24a/finkelshtein24a.pdf}, url = {https://proceedings.mlr.press/v235/finkelshtein24a.html}, abstract = {Graph neural networks are popular architectures for graph machine learning, based on iterative computation of node representations of an input graph through a series of invariant transformations. A large class of graph neural networks follow a standard message-passing paradigm: at every layer, each node state is updated based on an aggregate of messages from its neighborhood. In this work, we propose a novel framework for training graph neural networks, where every node is viewed as a player that can choose to either listen, broadcast, listen and broadcast, or to isolate. The standard message propagation scheme can then be viewed as a special case of this framework where every node listens and broadcasts to all neighbors. Our approach offers a more flexible and dynamic message-passing paradigm, where each node can determine its own strategy based on their state, effectively exploring the graph topology while learning. We provide a theoretical analysis of the new message-passing scheme which is further supported by an extensive empirical analysis on a synthetic and real-world datasets.} }
Endnote
%0 Conference Paper %T Cooperative Graph Neural Networks %A Ben Finkelshtein %A Xingyue Huang %A Michael M. Bronstein %A Ismail Ilkan Ceylan %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-finkelshtein24a %I PMLR %P 13633--13659 %U https://proceedings.mlr.press/v235/finkelshtein24a.html %V 235 %X Graph neural networks are popular architectures for graph machine learning, based on iterative computation of node representations of an input graph through a series of invariant transformations. A large class of graph neural networks follow a standard message-passing paradigm: at every layer, each node state is updated based on an aggregate of messages from its neighborhood. In this work, we propose a novel framework for training graph neural networks, where every node is viewed as a player that can choose to either listen, broadcast, listen and broadcast, or to isolate. The standard message propagation scheme can then be viewed as a special case of this framework where every node listens and broadcasts to all neighbors. Our approach offers a more flexible and dynamic message-passing paradigm, where each node can determine its own strategy based on their state, effectively exploring the graph topology while learning. We provide a theoretical analysis of the new message-passing scheme which is further supported by an extensive empirical analysis on a synthetic and real-world datasets.
APA
Finkelshtein, B., Huang, X., Bronstein, M.M. & Ceylan, I.I.. (2024). Cooperative Graph Neural Networks. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:13633-13659 Available from https://proceedings.mlr.press/v235/finkelshtein24a.html.

Related Material