Abstract:Neural-network quantum molecular dynamics (NNQMD) simulations based on machine learning are revolutionizing atomistic simulations of materials by providing quantum-mechanical accuracy but orders-of-magnitude faster, illustrated by ACM Gordon Bell prize (2020) and finalist (2021). State-of-the-art (SOTA) NNQMD model founded on group theory featuring rotational equivariance and local descriptors has provided much higher accuracy and speed than those models, thus named Allegro (meaning fast). On massively parallel supercomputers, however, it suffers a fidelity-scaling problem, where growing number of unphysical predictions of interatomic forces prohibits simulations involving larger numbers of atoms for longer times. Here, we solve this problem by combining the Allegro model with sharpness aware minimization (SAM) for enhancing the robustness of model through improved smoothness of the loss landscape. The resulting Allegro-Legato (meaning fast and "smooth") model was shown to elongate the time-to-failure $t_\textrm{failure}$, without sacrificing computational speed or accuracy. Specifically, Allegro-Legato exhibits much weaker dependence of timei-to-failure on the problem size, $t_{\textrm{failure}} \propto N^{-0.14}$ ($N$ is the number of atoms) compared to the SOTA Allegro model $\left(t_{\textrm{failure}} \propto N^{-0.29}\right)$, i.e., systematically delayed time-to-failure, thus allowing much larger and longer NNQMD simulations without failure. The model also exhibits excellent computational scalability and GPU acceleration on the Polaris supercomputer at Argonne Leadership Computing Facility. Such scalable, accurate, fast and robust NNQMD models will likely find broad applications in NNQMD simulations on emerging exaflop/s computers, with a specific example of accounting for nuclear quantum effects in the dynamics of ammonia.

Improved Optimization for the Neural-network Quantum States and Tests on the Chromium Dimer

A Non-stochastic Optimization Algorithm for Neural-network Quantum States

A Nonstochastic Optimization Algorithm for Neural-Network Quantum States

NNQS-Transformer: an Efficient and Scalable Neural Network Quantum States Approach for Ab initio Quantum Chemistry

Variational Quantum-Neural Hybrid Eigensolver.

Neural network backflow for ab-initio quantum chemistry

Empowering deep neural quantum states through efficient optimization

Neural Quantum States and Peaked Molecular Wave Functions: Curse or Blessing?

Quantum-Enhanced Neural Exchange-Correlation Functionals

Enhancing the Expressivity of Variational Neural, and Hardware-Efficient Quantum States Through Orbital Rotations

Neural-network-supported basis optimizer for the configuration interaction problem in quantum many-body clusters: Feasibility study and numerical proof

Variational Monte Carlo on a Budget -- Fine-tuning pre-trained Neural Wavefunctions

Quantum computing quantum Monte Carlo with hybrid tensor network for electronic structure calculations

Quantum Neural Network Inspired Hardware Adaptable Ansatz for Efficient Quantum Simulation of Chemical Systems

Allegro-Legato: Scalable, Fast, and Robust Neural-Network Quantum Molecular Dynamics via Sharpness-Aware Minimization

Variational optimization of the amplitude of neural-network quantum many-body ground states

Quantum Monte Carlo with reoptimized perturbatively selected configuration-interaction wave functions

Contextual Subspace Auxiliary-Field Quantum Monte Carlo: Improved bias with reduced quantum resources

SHARC-VQE: Simplified Hamiltonian Approach with Refinement and Correction enabled Variational Quantum Eigensolver for Molecular Simulation

Efficient molecular conformation generation with quantum-inspired algorithm