Abstract:Over the past decade, significant progress has been made in theoretical and experimental research in the field of chemical reaction dynamics, moving from triatomic reactions to larger polyatomic reactions. This has challenged the theoretical and computational approaches to polyatomic reaction dynamics in two major areas: the potential energy surface and the dynamics. Highly accurate potential energy surfaces are essential for achieving accurate dynamical information in quantum dynamics calculations. The increased number of degrees of freedom in larger systems poses a significant challenge to the accurate construction of potential energy surfaces. Recently, there has been substantial progress in the development of potential energy surfaces for polyatomic reactive systems. In this article, we review the recent developments made by our group in constructing highly accurately fitted potential energy surfaces for polyatomic reactive systems, based on a neural network approach. A key advantage of the neural network approach is its more faithful representation of the ab initio points. We recently proposed a systematic procedure, based on neural network fitting, for the construction of accurate potential energy surfaces with very small root mean square errors. Based on the neural network approach, we successfully developed potential energy surfaces for polyatomic reactions in the gas phase, including the reactive systems OH3, HOCO, and CH5, and the dissociation of gas-phase molecules on metal surfaces, such as H2O on the Cu(111) surface. These potential energy surfaces were fitted to an unprecedented level of accuracy, representing the most accurate potential energy surfaces calculated for these systems, and were rigorously tested using quantum dynamics calculations. The quantum dynamics calculations based on these potential energy surfaces produce accurate results, which are in good agreement with experiments. We have also proposed a new method for developing permutationally invariant potential energy surfaces, named fundamental-invariant neural networks. Mathematically, fundamental invariants are used to finitely generate the permutation-invariant polynomial ring; thus, fundamental-invariant neural networks can approximate any function to arbitrary accuracy. The use of fundamental invariants minimizes the size of the input permutation-invariant polynomials, which reduces the evaluation time for potential energy calculations, especially for polyatomic systems. Potential energy surfaces for OH3 and CH4 were constructed using fundamental-invariant neural networks, with their accuracies confirmed by full-dimensional quantum dynamics and bound-state calculations. These developments in the construction of highly accurate potential energy surfaces are expected to extend the theoretical study of reaction dynamics to larger and more complex systems.

Construction of Highly Accurate Machine Learning Potential Energy Surfaces for Excited-State Dynamics Simulations Based on Low-Level Data Sets

Deep Learning for Nonadiabatic Excited-State Dynamics.

Machine Learning Accelerates Precise Excited-State Potential Energy Surface Calculations on a Quantum Computer

Multi-Fidelity Machine Learning for Excited State Energies of Molecules

Inclusion of Machine Learning Kernel Ridge Regression Potential Energy Surfaces in On-the-Fly Nonadiabatic Molecular Dynamics Simulation.

Challenges for Machine Learning Force Fields in Reproducing Potential Energy Surfaces of Flexible Molecules

Direct Learning Hidden Excited State Interaction Patterns from Ab Initio Dynamics and Its Implication As Alternative Molecular Mechanism Models

Integrating Machine Learning with the Multilayer Energy-Based Fragment Method for Excited States of Large Systems

Efficient Construction of Excited-State Hessian Matrices with Machine Learning Accelerated Multilayer Energy-Based Fragment Method

Construction of Molecular Reactive Potential Energy Surfaces Based on Neural Networks

Machine-Learning-Assisted Free Energy Simulation of Solution-Phase and Enzyme Reactions

Machine Learning for Electronically Excited States of Molecules

Machine learning enables long time scale molecular photodynamics simulations

Charting electronic-state manifolds across molecules with multi-state learning and gap-driven dynamics via efficient and robust active learning

Large-Scale Atomic Simulation via Machine Learning Potentials Constructed by Global Potential Energy Surface Exploration

Neural Network Representations for Studying Gas‐Surface Reaction Dynamics: Beyond the Born‐Oppenheimer Static Surface Approximation †

Challenges and Opportunities for Machine Learning Potentials in Transition Path Sampling: Alanine Dipeptide and Azobenzene Studies

Exploring accurate potential energy surfaces via integrating variational quantum eigensovler with machine learning

Highly Accurately Fitted Potential Energy Surfaces for Polyatomic Reactive Systems

Machine learning accelerated photodynamics simulations

Transfer-Learned Potential Energy Surfaces: Towards Microsecond-Scale Molecular Dynamics Simulations in the Gas Phase at CCSD(T) Quality