Baking Symmetry into GFlowNets

TL;DR

This paper introduces methods to incorporate symmetry considerations into GFlowNets, aiming to improve training efficiency and diversity of generated objects by recognizing isomorphic actions during the sampling process.

Contribution

The paper proposes a novel approach to integrate symmetries into GFlowNets, addressing inefficiencies caused by isomorphic actions in current training pipelines.

Findings

Improved sample efficiency in synthetic experiments

Enhanced diversity of generated objects

Potential for more accurate flow functions

Abstract

GFlowNets have exhibited promising performance in generating diverse candidates with high rewards. These networks generate objects incrementally and aim to learn a policy that assigns probability of sampling objects in proportion to rewards. However, the current training pipelines of GFlowNets do not consider the presence of isomorphic actions, which are actions resulting in symmetric or isomorphic states. This lack of symmetry increases the amount of samples required for training GFlowNets and can result in inefficient and potentially incorrect flow functions. As a consequence, the reward and diversity of the generated objects decrease. In this study, our objective is to integrate symmetries into GFlowNets by identifying equivalent actions during the generation process. Experimental results using synthetic data demonstrate the promising performance of our proposed approaches.