Helping restricted Boltzmann machines with quantum-state representation by restoring symmetry

Abstract

The variational wave functions based on neural networks have recently started to be recognized as a powerful ansatz to represent quantum many-body states accurately. In order to show the usefulness of the method among all available numerical methods, it is imperative to investigate the performance in challenging many-body problems for which the exact solutions are not available. Here, we construct a variational wave function with one of the simplest neural networks, the restricted Boltzmann machine (RBM), and apply it to a fundamental but unsolved quantum spin Hamiltonian, the two-dimensional $J_1$-$J_2$ Heisenberg model on the square lattice. We supplement the RBM wave function with quantum-number projections, which restores the symmetry of the wave function and makes it possible to calculate excited states. Then, we perform a systematic investigation of the performance of the RBM. We show that, with the help of the symmetry, the RBM wave function achieves state-of-the-art accuracy both in ground-state and excited-state calculations. The study shows a practical guideline on how we achieve accuracy in a controlled manner.

10 pages, 7 figures, 3 tables, accepted for a special issue "Emerging Leaders 2020" in Journal of Physics: Condensed Matter

References in corpus (12)

• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• Restricted-Boltzmann-Machine Learning for Solving Strongly Correlated Quantum Systems
• Neural-Network Approach to Dissipative Quantum Many-Body Dynamics
• Variational Quantum Monte Carlo Method with a Neural-Network Ansatz for Open Quantum Systems
• Variational neural network ansatz for steady states in open quantum systems
• Constructing neural stationary states for open quantum many-body systems
• Variational Monte Carlo Method Combined with Quantum-Number Projection and Multi-Variable Optimization
• Solving the Bose-Hubbard model with machine learning
• Quantum Spin Liquid in Spin 1/2 J1-J2 Heisenberg Model on Square Lattice: Many-Variable Variational Monte Carlo Study Combined with Quantum-Number Projections
• Machine learning technique to find quantum many-body ground states of bosons on a lattice
• Quantum-number projection in the path-integral renormalization group method
• Variational wave functions for the spin-Peierls transition in the Su-Schrieffer-Heeger model with quantum phonons

Cited by in corpus (50)

• NetKet 3: Machine Learning Toolbox for Many-Body Quantum Systems
• Variational Benchmarks for Quantum Many-Body Problems
• Learning the ground state of a non-stoquastic quantum Hamiltonian in a rugged neural network landscape
• High-accuracy variational Monte Carlo for frustrated magnets with deep neural networks
• A simple linear algebra identity to optimize Large-Scale Neural Network Quantum States
• Optimizing Design Choices for Neural Quantum States
• Neural-network quantum states for many-body physics
• Purifying Deep Boltzmann Machines for Thermal Quantum States
• Determinant-free fermionic wave function using feed-forward neural networks
• Investigating Topological Order using Recurrent Neural Networks
• Expressive power of complex-valued restricted Boltzmann machines for solving non-stoquastic Hamiltonians
• Transformer Wave Function for two dimensional frustrated magnets: emergence of a Spin-Liquid Phase in the Shastry-Sutherland Model
• Many-Body Quantum States with Exact Conservation of Non-Abelian and Lattice Symmetries through Variational Monte Carlo
• Neural Network Evolution Strategy for Solving Quantum Sign Structures
• Highly resolved spectral functions of two-dimensional systems with neural quantum states
• jVMC: Versatile and performant variational Monte Carlo leveraging automated differentiation and GPU acceleration
• $2^{1296}$ Exponentially Complex Quantum Many-Body Simulation via Scalable Deep Learning Method
• Ground state search by local and sequential updates of neural network quantum states
• Spin-1/2 kagome Heisenberg antiferromagnet: Machine learning discovery of the spinon pair density wave ground state
• Update of $\mathcal{H}\Phi$: Newly added functions and methods in versions 2 and 3
• The quantum Gaussian process state: A kernel-inspired state with quantum support data
• Autoregressive neural Slater-Jastrow ansatz for variational Monte Carlo simulation
• Many-Qudit representation for the Travelling Salesman Problem Optimisation
• Compact Neural-network Quantum State representations of Jastrow and Stabilizer states
• Are queries and keys always relevant? A case study on Transformer wave functions
• Variational Quantum Eigensolver Ansatz for the $J_1$-$J_2$-model
• Impact of conditional modelling for a universal autoregressive quantum state
• Deep Neural Networks as Variational Solutions for Correlated Open Quantum Systems
• Neural-network Quantum States for Spin-1 systems: spin-basis and parameterization effects on compactness of representations
• Equivariant Variational Quantum Eigensolver to detect Phase Transitions through Energy Level Crossings
• Lee-Yang theory of quantum phase transitions with neural network quantum states
• Investigating Network Parameters in Neural-Network Quantum States
• Neural Network Quantum States analysis of the Shastry-Sutherland model
• Towards quantum gravity with neural networks: Solving the quantum Hamilton constraint of U(1) BF theory
• Time-dependent Neural Galerkin Method for Quantum Dynamics
• Boltzmann machines and quantum many-body problems
• Supervised Training of Neural-Network Quantum States for the Next Nearest Neighbor Ising model
• Quantum many-body solver using artificial neural networks and its applications to strongly correlated electron systems
• Machine Learning Wavefunction
• Broken-Symmetry Ground States of the Heisenberg model on the Pyrochlore Lattice
• Design principles of deep translationally-symmetric neural quantum states for frustrated magnets
• Efficient optimization and conceptual barriers in variational finite Projected Entangled-Pair States
• Improving neural network performance for solving quantum sign structure
• Kinetic samplers for neural quantum states
• Seeding neural network quantum states with tensor network states
• Accuracy of Restricted Boltzmann Machines for the one-dimensional $J_1-J_2$ Heisenberg model
• Group Convolutional Neural Network for the Low-Energy Spectrum in the Quantum Dimer Model
• Learning a compass spin model with neural network quantum states
• Comparing Symmetrized Determinant Neural Quantum States for the Hubbard Model
• Hybrid between biologically and quantum-inspired many-body states