Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders

Xue Yu, Muchen Li, Yan Leng, Renjie Liao
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:57507-57526, 2024.

Abstract

In network games, individuals interact strategically within network environments to maximize their utilities. However, obtaining network structures is challenging. In this work, we propose an unsupervised learning model, called data-dependent gated-prior graph variational autoencoder (GPGVAE), that infers the underlying latent interaction type (strategic complement vs. substitute) among individuals and the latent network structure based on their observed actions. Specially, we propose a spectral graph neural network (GNN) based encoder to predict the interaction type and a data-dependent gated prior that models network structures conditioned on the interaction type. We further propose a Transformer based mixture of Bernoulli encoder of network structures and a GNN based decoder of game actions. We systematically study the Monte Carlo gradient estimation methods and effectively train our model in a stage-wise fashion. Extensive experiments across various synthetic and real-world network games demonstrate that our model achieves state-of-the-art performances in inferring network structures and well captures interaction types.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-yu24f, title = {Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders}, author = {Yu, Xue and Li, Muchen and Leng, Yan and Liao, Renjie}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {57507--57526}, 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/yu24f/yu24f.pdf}, url = {https://proceedings.mlr.press/v235/yu24f.html}, abstract = {In network games, individuals interact strategically within network environments to maximize their utilities. However, obtaining network structures is challenging. In this work, we propose an unsupervised learning model, called data-dependent gated-prior graph variational autoencoder (GPGVAE), that infers the underlying latent interaction type (strategic complement vs. substitute) among individuals and the latent network structure based on their observed actions. Specially, we propose a spectral graph neural network (GNN) based encoder to predict the interaction type and a data-dependent gated prior that models network structures conditioned on the interaction type. We further propose a Transformer based mixture of Bernoulli encoder of network structures and a GNN based decoder of game actions. We systematically study the Monte Carlo gradient estimation methods and effectively train our model in a stage-wise fashion. Extensive experiments across various synthetic and real-world network games demonstrate that our model achieves state-of-the-art performances in inferring network structures and well captures interaction types.} }
Endnote
%0 Conference Paper %T Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders %A Xue Yu %A Muchen Li %A Yan Leng %A Renjie Liao %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-yu24f %I PMLR %P 57507--57526 %U https://proceedings.mlr.press/v235/yu24f.html %V 235 %X In network games, individuals interact strategically within network environments to maximize their utilities. However, obtaining network structures is challenging. In this work, we propose an unsupervised learning model, called data-dependent gated-prior graph variational autoencoder (GPGVAE), that infers the underlying latent interaction type (strategic complement vs. substitute) among individuals and the latent network structure based on their observed actions. Specially, we propose a spectral graph neural network (GNN) based encoder to predict the interaction type and a data-dependent gated prior that models network structures conditioned on the interaction type. We further propose a Transformer based mixture of Bernoulli encoder of network structures and a GNN based decoder of game actions. We systematically study the Monte Carlo gradient estimation methods and effectively train our model in a stage-wise fashion. Extensive experiments across various synthetic and real-world network games demonstrate that our model achieves state-of-the-art performances in inferring network structures and well captures interaction types.
APA
Yu, X., Li, M., Leng, Y. & Liao, R.. (2024). Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:57507-57526 Available from https://proceedings.mlr.press/v235/yu24f.html.

Related Material