Graph Diffusion Policy Optimization
This addresses a domain-specific challenge in graph generation for applications such as drug design, offering a novel method for a known bottleneck.
The paper tackles the problem of optimizing graph diffusion models for arbitrary objectives like non-differentiable ones in drug design, introducing Graph Diffusion Policy Optimization (GDPO) which achieves state-of-the-art performance in various graph generation tasks.
Recent research has made significant progress in optimizing diffusion models for downstream objectives, which is an important pursuit in fields such as graph generation for drug design. However, directly applying these models to graph presents challenges, resulting in suboptimal performance. This paper introduces graph diffusion policy optimization (GDPO), a novel approach to optimize graph diffusion models for arbitrary (e.g., non-differentiable) objectives using reinforcement learning. GDPO is based on an eager policy gradient tailored for graph diffusion models, developed through meticulous analysis and promising improved performance. Experimental results show that GDPO achieves state-of-the-art performance in various graph generation tasks with complex and diverse objectives. Code is available at https://github.com/sail-sg/GDPO.