Riemannian Denoising Model for Molecular Structure Optimization with Chemical Accuracy

TL;DR

This paper presents a Riemannian denoising model for molecular structure optimization that uses a physics-informed manifold to achieve chemical accuracy and outperform traditional Euclidean methods.

Contribution

It introduces a novel Riemannian manifold framework for molecular optimization, incorporating internal coordinates aligned with energetic properties for improved accuracy.

Findings

Achieves chemical accuracy with energy error below 1 kcal/mol.

Outperforms Euclidean-based denoising models on multiple datasets.

Demonstrates the effectiveness of physics-informed coordinates in molecular optimization.

Abstract

We introduce a framework for molecular structure optimization using denoising model on a physics-informed Riemannian manifold (R-DM). Unlike conventional approaches operating in Euclidean space, our method leverages a Riemannian metric that better aligns with molecular energy change, enabling more robust modeling of potential energy surfaces. By incorporating internal coordinates reflective of energetic properties, R-DM achieves chemical accuracy with an energy error below 1 kcal/mol. Comparative evaluations on QM9, QM7-X, and GEOM datasets demonstrate improvements in both structural and energetic accuracy, surpassing conventional Euclidean-based denoising models. This approach highlights the potential of physics-informed coordinates for tackling complex molecular optimization problems, with implications for tasks in computational chemistry and materials science.