Enabling ab initio geometry optimization of strongly correlated systems with transferable deep quantum Monte Carlo

2026-03-26

AI Summary

Accurately exploring potential energy surfaces involving bond breaking, bond formation, and large conformational changes in strongly correlated systems has long been hindered by the trade-off between accuracy and computational efficiency. This work proposes a novel approach that integrates transferable deep variational Monte Carlo (VMC) with Gaussian process regression, enabling low-cost estimation of energies, forces, and Hessians through continuous sampling of nuclear configurations during wavefunction optimization. For the first time in strongly correlated systems, the method achieves zero-shot chemical accuracy in geometry optimization and efficiently constructs noisy global potential energy surfaces. It successfully enables structural relaxations for both ground and excited states, transition state searches, and minimum energy path calculations.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Learning with Manifolds
• Reasoning under Uncertainty: Stochastic Optimization

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: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

A faithful description of chemical processes requires exploring extended regions of the molecular potential energy surface (PES), which remains challenging for strongly correlated systems. Transferable deep-learning variational Monte Carlo (VMC) offers a promising route by efficiently solving the electronic Schrödinger equation jointly across molecular geometries at consistently high accuracy, yet its stochastic nature renders direct exploration of molecular configuration space nontrivial. Here, we present a framework for highly accurate ab initio exploration of PESs that combines transferable deep-learning VMC with a cost-effective estimation of energies, forces, and Hessians. By continuously sampling nuclear configurations during VMC optimization of electronic wave functions, we obtain transferable descriptions that achieve zero-shot chemical accuracy within chemically relevant distributions of molecular geometries. Throughout the subsequent characterization of molecular configuration space, the PES is evaluated only sparsely, with local approximations constructed by estimating VMC energies and forces at sampled geometries and aggregating the resulting noisy data using Gaussian process regression. Our method enables accurate and efficient exploration of complex PES landscapes, including structure relaxation, transition-state searches, and minimum-energy pathways, for both ground and excited states. This opens the door to studying bond breaking, formation, and large structural rearrangements in systems with pronounced multi-reference character.

Problem

• strongly correlated systems
• potential energy surface
• geometry optimization
• ab initio
• chemical accuracy

Innovation

• transferable deep quantum Monte Carlo
• ab initio geometry optimization
• strongly correlated systems
• Gaussian process regression
• potential energy surface exploration