Sample Efficient Generative Molecular Optimization with Joint Self-Improvement

TL;DR

This paper introduces Joint Self-Improvement, a method combining a joint generative-predictive model and self-improving sampling to efficiently optimize molecules with limited evaluations, outperforming existing methods.

Contribution

It presents a novel joint model and sampling scheme that reduces distribution shift and improves sample efficiency in molecular optimization.

Findings

Outperforms state-of-the-art methods under limited evaluation budgets.

Effectively alleviates distribution shift in surrogate models.

Enhances sample efficiency in both offline and online benchmarks.

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.