Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Guanchen, Yuan, Chung-Yueh, Hegazy, Kareem, Blau, Samuel M., Head-Gordon, Teresa
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915881271951360
author Wu, Guanchen
Yuan, Chung-Yueh
Hegazy, Kareem
Blau, Samuel M.
Head-Gordon, Teresa
author_facet Wu, Guanchen
Yuan, Chung-Yueh
Hegazy, Kareem
Blau, Samuel M.
Head-Gordon, Teresa
contents The reliable determination of transition states (TSs) benefits from second-order information for robust convergence and validation, but the computational expense of Hessians prohibits their routine use in TS optimization. Here, we present a machine-learning-driven TS optimizer that directly predicts the leftmost Hessian eigenvector (LMHE), the critical mode that locally approximates the reaction coordinate encompassing the TS. We demonstrate that our LMHE optimizer recovers TS solutions at the same rate as full Hessian optimizers, and more robustly from degraded initial guess geometries, thereby eliminating the excessively long wall times characteristic of full-Hessian approaches and reducing total gradient evaluations compared to standard quasi-Newton methods. We further improve accuracy and robustness using uncertainty quantification for identifying occasional LMHE prediction failures, that then falls back to a full Hessian update from the machine learned potential at that optimization step, avoiding expensive active learning. Overall our methodology and semi-automated workflow delivers second-order stability at first-order computational expense to provide a highly efficient engine for high-throughput reaction discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2603_21323
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding
Wu, Guanchen
Yuan, Chung-Yueh
Hegazy, Kareem
Blau, Samuel M.
Head-Gordon, Teresa
Chemical Physics
The reliable determination of transition states (TSs) benefits from second-order information for robust convergence and validation, but the computational expense of Hessians prohibits their routine use in TS optimization. Here, we present a machine-learning-driven TS optimizer that directly predicts the leftmost Hessian eigenvector (LMHE), the critical mode that locally approximates the reaction coordinate encompassing the TS. We demonstrate that our LMHE optimizer recovers TS solutions at the same rate as full Hessian optimizers, and more robustly from degraded initial guess geometries, thereby eliminating the excessively long wall times characteristic of full-Hessian approaches and reducing total gradient evaluations compared to standard quasi-Newton methods. We further improve accuracy and robustness using uncertainty quantification for identifying occasional LMHE prediction failures, that then falls back to a full Hessian update from the machine learned potential at that optimization step, avoiding expensive active learning. Overall our methodology and semi-automated workflow delivers second-order stability at first-order computational expense to provide a highly efficient engine for high-throughput reaction discovery.
title Machine-Learned Leftmost Hessian Eigenvectors for Robust Transition State Finding
topic Chemical Physics
url https://arxiv.org/abs/2603.21323