Implicit Neural Representations for Chemical Reaction Paths

Abstract

We show that neural networks can be optimized to represent minimum energy paths as continuous functions, offering a flexible alternative to discrete path-search methods such as Nudged Elastic Band (NEB). Our approach parameterizes reaction paths with a network trained on a loss function that discards tangential energy gradients and enables instant estimation of the transition state. We first validate the method on two-dimensional potentials and then demonstrate its advantages over NEB on challenging atomistic systems where (i) poor initial guesses yield unphysical paths, (ii) multiple competing paths exist, or (iii) the reaction follows a complex multi-step mechanism. Results highlight the versatility of the method: for instance, a simple adjustment to the sampling strategy during optimization can help escape local-minimum solutions. Finally, in a low-dimensional setting, we demonstrate that a single neural network can learn from existing paths and generalize to unseen systems, showing promise for a universal reaction path representation.

Published in the Journal of Chemical Physics

References in corpus (14)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Quantum ESPRESSO toward the exascale
• String Method for the Study of Rare Events
• Improved initial guess for minimum energy path calculations
• Global transition path search for dislocation formation in Ge on Si(001)
• A Doubly Nudged Elastic Band Method for Finding Transition States
• Materials Cloud, a platform for open computational science
• MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields
• Implicit Neural Representations with Periodic Activation Functions
• Averaging Weights Leads to Wider Optima and Better Generalization
• E(n) Equivariant Graph Neural Networks
• Minimum Energy Paths and Transition States by Curve Optimization
• A preconditioning scheme for Minimum Energy Path finding methods
• Doob's Lagrangian: A Sample-Efficient Variational Approach to Transition Path Sampling