Role of stochastic noise and generalization error in the time propagation of neural-network quantum states

Abstract

Neural-network quantum states (NQS) have been shown to be a suitable variational ansatz to simulate out-of-equilibrium dynamics in two-dimensional systems using time-dependent variational Monte Carlo (t-VMC). In particular, stable and accurate time propagation over long time scales has been observed in the square-lattice Heisenberg model using the Restricted Boltzmann machine architecture. However, achieving similar performance in other systems has proven to be more challenging. In this article, we focus on the two-leg Heisenberg ladder driven out of equilibrium by a pulsed excitation as a benchmark system. We demonstrate that unmitigated noise is strongly amplified by the nonlinear equations of motion for the network parameters, which causes numerical instabilities in the time evolution. As a consequence, the achievable accuracy of the simulated dynamics is a result of the interplay between network expressiveness and measures required to remedy these instabilities. We show that stability can be greatly improved by appropriate choice of regularization. This is particularly useful as tuning of the regularization typically imposes no additional computational cost. Inspired by machine learning practice, we propose a validation-set based diagnostic tool to help determining optimal regularization hyperparameters for t-VMC based propagation schemes. For our benchmark, we show that stable and accurate time propagation can be achieved in regimes of sufficiently regularized variational dynamics.

arXiv comment27 pages, 13 figures

References in corpus (53)

• Area laws for the entanglement entropy - a review
• Machine learning and the physical sciences
• QuTiP 2: A Python framework for the dynamics of open quantum systems
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Quantum fluids of light
• Surprises on the Way from 1D to 2D Quantum Magnets: the Novel Ladder Materials
• Machine learning phases of matter
• Quantum many-body systems out of equilibrium
• Many-body quantum state tomography with neural networks
• Nonthermal pathways to ultrafast control in quantum materials
• Quantum Natural Gradient
• Controlling many-body dynamics with driven quantum scars in Rydberg atom arrays
• Quantum Entanglement in Neural Network States
• Spin-ladders with spin gaps: A description of a class of cuprates
• Restricted-Boltzmann-Machine Learning for Solving Strongly Correlated Quantum Systems
• Reinforcement Learning in Different Phases of Quantum Control
• Neural-Network Approach to Dissipative Quantum Many-Body Dynamics
• Deep autoregressive models for the efficient variational simulation of many-body quantum systems
• Symmetries and many-body excited states with neural-network quantum states
• Fermionic neural-network states for ab-initio electronic structure
• Study of the Two-Dimensional Frustrated J1-J2 Model with Neural Network Quantum States
• Quantum many-body dynamics in two dimensions with artificial neural networks
• Identifying Quantum Phase Transitions using Artificial Neural Networks on Experimental Data
• Variational neural network ansatz for steady states in open quantum systems
• Reinforcement Learning with Neural Networks for Quantum Feedback
• Neural-Network Quantum States, String-Bond States, and Chiral Topological States
• Localization and Glassy Dynamics Of Many-Body Quantum Systems
• Constructing neural stationary states for open quantum many-body systems
• Latent Space Purification via Neural Density Operators
• Neural Quantum States of frustrated magnets: generalization and sign structure
• NetKet: A Machine Learning Toolkit for Many-Body Quantum Systems
• Classifying Snapshots of the Doped Hubbard Model with Machine Learning
• Dynamics of quantum information
• Solving frustrated quantum many-particle models with convolutional neural networks
• Restricted Boltzmann Machines for Quantum States with Nonabelian or Anyonic Symmetries
• Chiral Topological Phases from Artificial Neural Networks
• Quenches near Ising quantum criticality as a challenge for artificial neural networks
• Neural network wave functions and the sign problem
• Learning the ground state of a non-stoquastic quantum Hamiltonian in a rugged neural network landscape
• Geometry of learning neural quantum states
• Laser-driven quantum magnonics and THz dynamics of the order parameter in antiferromagnets
• Investigating ultrafast quantum magnetism with machine learning
• A machine learning approach to dynamical properties of quantum many-body systems
• Neural-network states for the classical simulation of quantum computing
• Generalized Transfer Matrix States from Artificial Neural Networks
• Scaling of neural-network quantum states for time evolution
• Expressive power of complex-valued restricted Boltzmann machines for solving non-stoquastic Hamiltonians
• Correlation-Enhanced Neural Networks as Interpretable Variational Quantum States
• Real time evolution with neural-network quantum states
• Supermagnonic propagation in two-dimensional antiferromagnets
• Ground state phase diagram of the one-dimensional Bose-Hubbard model from restricted Boltzmann machines
• Sign Rules for Anisotropic Quantum Spin Systems
• Marshall's sign rule and DMRG acceleration

Cited by in corpus (15)

• NetKet 3: Machine Learning Toolbox for Many-Body Quantum Systems
• Optimizing Design Choices for Neural Quantum States
• Scaling of neural-network quantum states for time evolution
• Dynamics with autoregressive neural quantum states: application to critical quench dynamics
• Variational quantum dynamics of two-dimensional rotor models
• jVMC: Versatile and performant variational Monte Carlo leveraging automated differentiation and GPU acceleration
• Spin-1/2 kagome Heisenberg antiferromagnet: Machine learning discovery of the spinon pair density wave ground state
• Cavity engineering of Hubbard $U$ via phonon polaritons
• Impact of conditional modelling for a universal autoregressive quantum state
• Spectroscopy of two-dimensional interacting lattice electrons using symmetry-aware neural backflow transformations
• An Empirical Study of Quantum Dynamics as a Ground State Problem with Neural Quantum States
• Learning topological defects formation with neural networks in a quantum phase transition
• Correcting and extending Trotterized quantum many-body dynamics
• Predicting sampling advantage of stochastic Ising Machines for Quantum Simulations
• Adaptive quantum dynamics with the time-dependent variational Monte Carlo method