Expected Flow Networks in Stochastic Environments and Two-Player Zero-Sum Games
Abstract
Generative flow networks (GFlowNets) are sequential sampling models trained to match a given distribution. GFlowNets have been successfully applied to various structured object generation tasks, sampling a diverse set of high-reward objects quickly. We propose expected flow networks (EFlowNets), which extend GFlowNets to stochastic environments. We show that EFlowNets outperform other GFlowNet formulations in stochastic tasks such as protein design. We then extend the concept of EFlowNets to adversarial environments, proposing adversarial flow networks (AFlowNets) for two-player zero-sum games. We show that AFlowNets learn to find above 80% of optimal moves in Connect-4 via self-play and outperform AlphaZero in tournaments. Code: https://github.com/GFNOrg/AdversarialFlowNetworks.
Cite
Text
Jiralerspong et al. "Expected Flow Networks in Stochastic Environments and Two-Player Zero-Sum Games." International Conference on Learning Representations, 2024.Markdown
[Jiralerspong et al. "Expected Flow Networks in Stochastic Environments and Two-Player Zero-Sum Games." International Conference on Learning Representations, 2024.](https://mlanthology.org/iclr/2024/jiralerspong2024iclr-expected/)BibTeX
@inproceedings{jiralerspong2024iclr-expected,
title = {{Expected Flow Networks in Stochastic Environments and Two-Player Zero-Sum Games}},
author = {Jiralerspong, Marco and Sun, Bilun and Vucetic, Danilo and Zhang, Tianyu and Bengio, Yoshua and Gidel, Gauthier and Malkin, Nikolay},
booktitle = {International Conference on Learning Representations},
year = {2024},
url = {https://mlanthology.org/iclr/2024/jiralerspong2024iclr-expected/}
}