Efficient Nudged Elastic Band Method using Neural Network Bayesian Algorithm Execution

2025-12-16

AI Summary

Traditional Nudged Elastic Band (NEB) methods incur prohibitive computational cost when searching for minimum energy paths (MEPs) on high-dimensional potential energy surfaces (PES), especially beyond 100 dimensions. To address this, we propose a Neural Network–Bayesian Active eXploration (NN-BAX) framework that jointly optimizes PES modeling and MEP search via Bayesian optimization, deep neural network surrogate modeling, and MEP-guided sequential active sampling. Compared to standard NEB, NN-BAX reduces energy and force evaluations by 90–99% on benchmark systems—including Lennard-Jones and embedded-atom method (EAM) potentials—while preserving MEP accuracy within negligible error margins. Computational time is reduced from weeks to hours or days. Crucially, NN-BAX enables, for the first time, high-accuracy and high-efficiency MEP computation on PES with over 100 degrees of freedom.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Bayesian Learning
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

The discovery of a minimum energy pathway (MEP) between metastable states is crucial for scientific tasks including catalyst and biomolecular design. However, the standard nudged elastic band (NEB) algorithm requires hundreds to tens of thousands of compute-intensive simulations, making applications to complex systems prohibitively expensive. We introduce Neural Network Bayesian Algorithm Execution (NN-BAX), a framework that jointly learns the energy landscape and the MEP. NN-BAX sequentially fine-tunes a foundation model by actively selecting samples targeted at improving the MEP. Tested on Lennard-Jones and Embedded Atom Method systems, our approach achieves a one to two order of magnitude reduction in energy and force evaluations with negligible loss in MEP accuracy and demonstrates scalability to >100-dimensional systems. This work is therefore a promising step towards removing the computational barrier for MEP discovery in scientifically relevant systems, suggesting that weeks-long calculations may be achieved in hours or days with minimal loss in accuracy.

Problem

• Reduces computational cost of finding minimum energy pathways
• Accelerates catalyst and biomolecular design simulations
• Enables scalable MEP discovery in high-dimensional systems

Innovation

• Neural network Bayesian algorithm execution framework
• Active selection of samples for fine-tuning foundation model
• Reduces energy evaluations by orders of magnitude