Sample Efficient Generative Molecular Optimization with Joint Self-Improvement

2026-02-11

AI Summary

This work proposes a joint self-improvement framework to address the challenges of low sample efficiency and distributional shift that degrade surrogate model performance in generative molecular optimization. By coupling generative and predictive models, the approach mitigates distributional shift, while a novel self-improvement sampling strategy enhances sample utilization efficiency. During inference, the framework efficiently generates high-quality molecules with desirable properties, substantially reducing reliance on costly property evaluations. Experimental results demonstrate that, under limited evaluation budgets, the method significantly outperforms current state-of-the-art approaches on both offline and online molecular optimization benchmarks, confirming its effectiveness and practical utility.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Optimization
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Generative molecular optimization aims to design molecules with properties surpassing those of existing compounds. However, such candidates are rare and expensive to evaluate, yielding sample efficiency essential. Additionally, surrogate models introduced to predict molecule evaluations, suffer from distribution shift as optimization drives candidates increasingly out-of-distribution. To address these challenges, we introduce Joint Self-Improvement, which benefits from (i) a joint generative-predictive model and (ii) a self-improving sampling scheme. The former aligns the generator with the surrogate, alleviating distribution shift, while the latter biases the generative part of the joint model using the predictive one to efficiently generate optimized molecules at inference-time. Experiments across offline and online molecular optimization benchmarks demonstrate that Joint Self-Improvement outperforms state-of-the-art methods under limited evaluation budgets.

Problem

• sample efficiency
• generative molecular optimization
• distribution shift
• surrogate models
• molecular design

Innovation

• generative molecular optimization
• sample efficiency
• joint generative-predictive model
• distribution shift
• self-improving sampling