Genetic-guided GFlowNets for Sample Efficient Molecular Optimization

2024-02-05

AI Summary

Molecular optimization in drug and materials discovery suffers from expensive reward function evaluations and low sample efficiency. Method: We propose a novel domain-knowledge-enhanced GFlowNets framework that distills prior knowledge from genetic algorithms—specifically selection, crossover, and mutation operators—into the GFlowNets generative policy. This yields an off-policy, amortized graph-structured molecular search framework, integrating graph neural networks with flow matching-based training. Contribution/Results: Our method achieves state-of-the-art performance on standard molecular optimization benchmarks. In SARS-CoV-2 inhibitor design, it reduces reward function calls by over 50%, significantly improving optimization efficiency and computational scalability. Crucially, it successfully embeds the inductive biases of classical evolutionary algorithms into a deep generative model, unifying high sample efficiency with strong generalization capability.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Optimization
• Constraint Satisfaction and Optimization: Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

The challenge of discovering new molecules with desired properties is crucial in domains like drug discovery and material design. Recent advances in deep learning-based generative methods have shown promise but face the issue of sample efficiency due to the computational expense of evaluating the reward function. This paper proposes a novel algorithm for sample-efficient molecular optimization by distilling a powerful genetic algorithm into deep generative policy using GFlowNets training, the off-policy method for amortized inference. This approach enables the deep generative policy to learn from domain knowledge, which has been explicitly integrated into the genetic algorithm. Our method achieves state-of-the-art performance in the official molecular optimization benchmark, significantly outperforming previous methods. It also demonstrates effectiveness in designing inhibitors against SARS-CoV-2 with substantially fewer reward calls.

Problem

• Molecular Optimization
• Computational Efficiency
• Drug Discovery

Innovation

• GFlowNets
• Genetic Algorithm
• Deep Learning