Expected flow networks in stochastic environments and two-player zero-sum games
cs.LG, cs.GT
Submitted: 2023-10-04
Updated: 2026-08-31
Comments: ICLR 2024; code: https://github.com/GFNOrg/AdversarialFlowNetworks
Code: https://github.com/GFNOrg/AdversarialFlowNetworks
License: http://creativecommons.org/licenses/by/4.0/
The gist: Generative flow networks (GFlowNets) are sequential sampling models trained to match a given distribution.
Terminology
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.
Sources
- Joint Bayesian Inference of Graphical Structure and Parameters with a Single Generative Flow Network
- Thompson sampling for improved exploration in GFlowNets
- A-NeSI: A Scalable Approximate Method for Probabilistic Neurosymbolic Inference
- Distributional GFlowNets with Quantile Flows
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks