Symmetry-Aware GFlowNets

TL;DR

Symmetry-Aware GFlowNets (SA-GFN) improve graph sampling by correcting symmetry-induced biases through reward scaling, leading to unbiased, diverse, and high-quality graph generation without explicit transition calculations.

Contribution

The paper introduces SA-GFN, a novel method that incorporates symmetry corrections into GFlowNets via reward scaling, eliminating the need for explicit transition probability computations.

Findings

SA-GFN achieves unbiased graph sampling.

SA-GFN enhances diversity in generated graphs.

SA-GFN produces high-reward graphs closely matching target distribution.

Abstract

Generative Flow Networks (GFlowNets) offer a powerful framework for sampling graphs in proportion to their rewards. However, existing approaches suffer from systematic biases due to inaccuracies in state transition probability computations. These biases, rooted in the inherent symmetries of graphs, impact both atom-based and fragment-based generation schemes. To address this challenge, we introduce Symmetry-Aware GFlowNets (SA-GFN), a method that incorporates symmetry corrections into the learning process through reward scaling. By integrating bias correction directly into the reward structure, SA-GFN eliminates the need for explicit state transition computations. Empirical results show that SA-GFN enables unbiased sampling while enhancing diversity and consistently generating high-reward graphs that closely match the target distribution.