Amortized Molecular Optimization via Group Relative Policy Optimization

• URL: http://arxiv.org/abs/2602.12162v1
• Date: Thu, 12 Feb 2026 16:43:59 GMT
• Title: Amortized Molecular Optimization via Group Relative Policy Optimization
• Abstract summary: We introduce GRXForm, adapting a pre-trained Graph Transformer model that optimize molecules via sequential atom-and-bond additions.<n> GRXForm generalizes to out-of-distribution molecular scaffolds without inference-time oracle calls or refinement.
• Score: 38.75198089892667
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Molecular design encompasses tasks ranging from de-novo design to structural alteration of given molecules or fragments. For the latter, state-of-the-art methods predominantly function as "Instance Optimizers'', expending significant compute restarting the search for every input structure. While model-based approaches theoretically offer amortized efficiency by learning a policy transferable to unseen structures, existing methods struggle to generalize. We identify a key failure mode: the high variance arising from the heterogeneous difficulty of distinct starting structures. To address this, we introduce GRXForm, adapting a pre-trained Graph Transformer model that optimizes molecules via sequential atom-and-bond additions. We employ Group Relative Policy Optimization (GRPO) for goal-directed fine-tuning to mitigate variance by normalizing rewards relative to the starting structure. Empirically, GRXForm generalizes to out-of-distribution molecular scaffolds without inference-time oracle calls or refinement, achieving scores in multi-objective optimization competitive with leading instance optimizers.

Related papers

• Constrained Flow Optimization via Sequential Fine Tuning for Molecular Design [31.59334876687123]
  We present a framework for Constrained Generative Optimization (CFO)<n>CFO balances reward and constraint satisfaction by reducing the original problem to sequential fine-tuning.<n>We provide convergence guarantees for constrained generative optimization and constrained generation via CFO.
  arXiv  Detail & Related papers  (2026-05-28T22:02:57Z)
• Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Compute [64.59043351161004]
  We propose Proteina-Complexa, a novel fully atomistic binder generation method unifying both paradigms.<n>Proteina-Complexa sets a new state of the art in computational binder design benchmarks.<n>We also demonstrate interface hydrogen bond optimization, fold class-guided binder generation, and extensions to small molecule targets and enzyme design tasks.
  arXiv  Detail & Related papers  (2026-03-30T01:54:03Z)
• Symplectic Optimization on Gaussian States [0.0]
  We introduce a symplectic optimization framework to solve the bosonic ground-state problem.<n>The framework provides a foundation for large-scale approximate treatments of weakly non-quadratic interactions.
  arXiv  Detail & Related papers  (2026-01-28T18:31:50Z)
• Reward driven discovery of the optimal microstructure representations with invariant variational autoencoders [0.015295722752489374]
  Variational Autoencoders (VAEs) provide a powerful means of constructing such low-dimensional representations.<n>VAEs are often optimized through trial-and-error and empirical analysis.<n>We investigated reward-based strategies for evaluating latent space representations.
  arXiv  Detail & Related papers  (2025-09-30T20:15:42Z)
• Bayesian Optimization for Molecules Should Be Pareto-Aware [6.877358955271652]
  Multi-objective Bayesian optimization (MOBO) provides a principled framework for navigating trade-offs in molecular design.<n>We benchmark a simple MOBO strategy against a simple fixed-weight scalarized baseline using Expected Improvement (EI)<n>Our results show that even strong deterministic instantiations can underperform in low-data regimes.
  arXiv  Detail & Related papers  (2025-07-18T07:12:19Z)
• Return of the Latent Space COWBOYS: Re-thinking the use of VAEs for Bayesian Optimisation of Structured Spaces [13.38402522324075]
  We propose a decoupled approach that trains a generative model and a Gaussian Process (GP) surrogate separately, then combines them via a simple yet principled Bayesian update rule.<n>We show that our decoupled approach improves our ability to identify high-potential candidates in molecular optimisation problems under constrained evaluation budgets.
  arXiv  Detail & Related papers  (2025-07-05T05:53:04Z)
• Cliqueformer: Model-Based Optimization with Structured Transformers [102.55764949282906]
  Large neural networks excel at prediction tasks, but their application to design problems, such as protein engineering or materials discovery, requires solving offline model-based optimization (MBO) problems.<n>We present Cliqueformer, a transformer-based architecture that learns the black-box function's structure through functional graphical models (FGM)<n>Across various domains, including chemical and genetic design tasks, Cliqueformer demonstrates superior performance compared to existing methods.
  arXiv  Detail & Related papers  (2024-10-17T00:35:47Z)
• DecompOpt: Controllable and Decomposed Diffusion Models for Structure-based Molecular Optimization [49.85944390503957]
  DecompOpt is a structure-based molecular optimization method based on a controllable and diffusion model. We show that DecompOpt can efficiently generate molecules with improved properties than strong de novo baselines.
  arXiv  Detail & Related papers  (2024-03-07T02:53:40Z)
• Backpropagation of Unrolled Solvers with Folded Optimization [55.04219793298687]
  The integration of constrained optimization models as components in deep networks has led to promising advances on many specialized learning tasks. One typical strategy is algorithm unrolling, which relies on automatic differentiation through the operations of an iterative solver. This paper provides theoretical insights into the backward pass of unrolled optimization, leading to a system for generating efficiently solvable analytical models of backpropagation.
  arXiv  Detail & Related papers  (2023-01-28T01:50:42Z)
• When to Update Your Model: Constrained Model-based Reinforcement Learning [50.74369835934703]
  We propose a novel and general theoretical scheme for a non-decreasing performance guarantee of model-based RL (MBRL) Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns.
  arXiv  Detail & Related papers  (2022-10-15T17:57:43Z)
• EBM-Fold: Fully-Differentiable Protein Folding Powered by Energy-based Models [53.17320541056843]
  We propose a fully-differentiable approach for protein structure optimization, guided by a data-driven generative network. Our EBM-Fold approach can efficiently produce high-quality decoys, compared against traditional Rosetta-based structure optimization routines.
  arXiv  Detail & Related papers  (2021-05-11T03:40:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.