RLSynC: Offline-Online Reinforcement Learning for Synthon Completion
This addresses synthesis planning in chemistry, offering a novel approach for improving retrosynthesis accuracy, though it is incremental within semi-template-based methods.
The paper tackled the problem of synthon completion in retrosynthesis by developing RLSynC, an offline-online reinforcement learning method that uses multiple agents and a forward synthesis model, resulting in improvements as high as 14.9% over state-of-the-art methods.
Retrosynthesis is the process of determining the set of reactant molecules that can react to form a desired product. Semi-template-based retrosynthesis methods, which imitate the reverse logic of synthesis reactions, first predict the reaction centers in the products, and then complete the resulting synthons back into reactants. We develop a new offline-online reinforcement learning method RLSynC for synthon completion in semi-template-based methods. RLSynC assigns one agent to each synthon, all of which complete the synthons by conducting actions step by step in a synchronized fashion. RLSynC learns the policy from both offline training episodes and online interactions, which allows RLSynC to explore new reaction spaces. RLSynC uses a standalone forward synthesis model to evaluate the likelihood of the predicted reactants in synthesizing a product, and thus guides the action search. Our results demonstrate that RLSynC can outperform state-of-the-art synthon completion methods with improvements as high as 14.9%, highlighting its potential in synthesis planning.