cs.LGFeb 12, 2026

Amortized Molecular Optimization via Group Relative Policy Optimization

Authors: Muhammad bin Javaid, Hasham Hussain, Ashima Khanna-Reiter, Berke Kişin, Jonathan Pirnay, Alexander Mitsos, Dominik G. Grimm, Martin Grohe

Organizations: RWTH Aachen University, Department of Computer Science, Aachen, Germany · Alfred E. Tiefenbacher (GmbH & Co. KG), Hamburg, Germany · Technical University of Munich, TUM Campus Straubing for Biotechnology and Sustainability · University of Applied Sciences Weihenstephan-Triesdorf, Bioinformatics · 

Abstract

In structurally constrained molecular optimization, state-of-the-art methods treat every new input structure as a separate instance, restarting an expensive oracle-driven search from scratch and scaling poorly to settings with many starting structures or expensive oracles. While amortized approaches that learn a transferable policy could in principle remove this bottleneck, existing methods struggle to generalize to diverse structural constraints at inference time. We present AMORTIX, a Graph Transformer model that natively supports such constraints, optimizing molecular structures in a single rollout with zero inference-time oracle calls. A central challenge for amortized training in this domain is that optimization difficulty varies drastically across starting structures. We address this by centering rewards within groups of completions that share the same starting structure. We evaluate on goal-directed scaffold decoration, including single- and multi-target kinase inhibitor design, and on a few-shot prodrug case study. AMORTIX outperforms both amortized baselines and search-based instance optimizers on kinase scaffold decoration, and ablations across 23 objectives show that group-relative centering improves over REINFORCE and PPO and is substantially more consistent across scaffold splits. The prodrug case study further demonstrates transfer of learned modification rules to unseen drug structures.

Figures & tables

Appendix figures & tables22 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Reference-guided Policy Optimization for Molecular Optimization via LLM Reasoning

    Mar 6, 2026Xuan Li, Zhanke Zhou, Zongze Li +4Molecular OptimizationLarge Language Model-Guided Optimization

  2. MARCO: Multi-Round Agentic Reinforcement for Conditional Molecular Optimization

    Sep 29, 2026Shicheng Fang, Yuxin Wang, Zhuo Yang +6Molecular OptimizationRL for Language Models