Neural Quantum States of frustrated magnets: generalization and sign structure

Abstract

Neural quantum states (NQS) attract a lot of attention due to their potential to serve as a very expressive variational ansatz for quantum many-body systems. Here we study the main factors governing the applicability of NQS to frustrated magnets by training neural networks to approximate ground states of several moderately-sized Hamiltonians using the corresponding wavefunction structure on a small subset of the Hilbert space basis as training dataset. We notice that generalization quality, i.e. the ability to learn from a limited number of samples and correctly approximate the target state on the rest of the space, drops abruptly when frustration is increased. We also show that learning the sign structure is considerably more difficult than learning amplitudes. Finally, we conclude that the main issue to be addressed at this stage, in order to use the method of NQS for simulating realistic models, is that of generalization rather than expressibility.

arXiv comment9+6 pages, 5+7 figures, v3: added results for larger systems, improved discussion, added new references

References in corpus (45)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Machine learning and the physical sciences
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• A Practical Introduction to Tensor Networks: Matrix Product States and Projected Entangled Pair States
• Machine learning phases of matter
• Many-body quantum state tomography with neural networks
• Learning phase transitions by confusion
• Matrix product states represent ground states faithfully
• By-passing the Kohn-Sham equations with machine learning
• Quantum Entanglement in Neural Network States
• Machine learning quantum phases of matter beyond the fermion sign problem
• Generalized Lanczos Algorithm for Variational Quantum Monte Carlo
• Spin Liquid Ground State of the Spin-1/2 Square $J_1$-$J_2$ Heisenberg Model
• Discovering Phases, Phase Transitions and Crossovers through Unsupervised Machine Learning: A critical examination
• Efficient Representation of Quantum Many-body States with Deep Neural Networks
• Restricted-Boltzmann-Machine Learning for Solving Strongly Correlated Quantum Systems
• Neural-Network Approach to Dissipative Quantum Many-Body Dynamics
• Symmetries and many-body excited states with neural-network quantum states
• Study of the Two-Dimensional Frustrated J1-J2 Model with Neural Network Quantum States
• Weak binding between two aromatic rings: feeling the van der Waals attraction by quantum Monte Carlo methods
• Variational neural network ansatz for steady states in open quantum systems
• Correlated geminal wave function for molecules: an efficient resonating valence bond approach
• Computing Nonvacuous Generalization Bounds for Deep (Stochastic) Neural Networks with Many More Parameters than Training Data
• Approximating quantum many-body wave-functions using artificial neural networks
• Two spin liquid phases in the spatially anisotropic triangular Heisenberg model
• Magnetic order and disorder in the frustrated quantum Heisenberg antiferromagnet in two dimensions
• Quantum Entanglement in Deep Learning Architectures
• Energy and variance optimization of many body wave functions
• Variational description of Mott insulators
• Many-body localization and delocalization in large quantum chains
• Backflow Transformations via Neural Networks for Quantum Many-Body Wave-Functions
• The spin-1/2 J1-J2 Heisenberg antiferromagnet on the square lattice: Exact diagonalization for N=40 spins
• Constructing exact representations of quantum many-body systems with deep neural networks
• Latent Space Purification via Neural Density Operators
• Subdominant Dense Clusters Allow for Simple Learning and High Computational Performance in Neural Networks with Discrete Synapses
• Machine learning technique to find quantum many-body ground states of bosons on a lattice
• Solving frustrated quantum many-particle models with convolutional neural networks
• Neural Gutzwiller-projected variational wave functions
• Optimized Jastrow-Slater wave functions for ground and excited states: Application to the lowest states of ethene
• Spatially Anisotropic Heisenberg Kagome Antiferromagnet
• Transfer learning for scalability of neural-network quantum states
• The learnability scaling of quantum states: restricted Boltzmann machines
• Spin-1/2 Heisenberg antiferromagnet on an anisotropic kagome lattice
• Wavefunction positivization via automatic differentiation
• Neural-network quantum states at finite temperature

Cited by in corpus (83)

• Dirac-type nodal spin liquid revealed by refined quantum many-body solver using neural-network wave function, correlation ratio, and level spectroscopy
• Transformer variational wave functions for frustrated quantum spin systems
• Neural network wave functions and the sign problem
• Neural network quantum state tomography in a two-qubit experiment
• Empowering deep neural quantum states through efficient optimization
• Autoregressive Transformer Neural Network for Simulating Open Quantum Systems via a Probabilistic Formulation
• Variational Quantum Eigensolver for Frustrated Quantum Systems
• Helping restricted Boltzmann machines with quantum-state representation by restoring symmetry
• Learning the ground state of a non-stoquastic quantum Hamiltonian in a rugged neural network landscape
• Variational Quantum-Neural Hybrid Eigensolver
• High-accuracy variational Monte Carlo for frustrated magnets with deep neural networks
• Phases of two-dimensional spinless lattice fermions with first-quantized deep neural-network quantum states
• Transfer learning for scalability of neural-network quantum states
• Optimizing Design Choices for Neural Quantum States
• Speeding up Learning Quantum States through Group Equivariant Convolutional Quantum Ansätze
• Deep learning of quantum entanglement from incomplete measurements
• Learning the Ising Model with Generative Neural Networks
• Hybrid convolutional neural network and PEPS wave functions for quantum many-particle states
• Neural-network quantum states for many-body physics
• Purifying Deep Boltzmann Machines for Thermal Quantum States
• Direct sampling of projected entangled-pair states
• From Tensor Network Quantum States to Tensorial Recurrent Neural Networks
• Scaling of neural-network quantum states for time evolution
• Determinant-free fermionic wave function using feed-forward neural networks
• Role of stochastic noise and generalization error in the time propagation of neural-network quantum states
• Intrinsic dimension of path integrals: data mining quantum criticality and emergent simplicity
• Expressive power of complex-valued restricted Boltzmann machines for solving non-stoquastic Hamiltonians
• Machine Learning Quantum States -- Extensions to Fermion-Boson Coupled Systems and Excited-State Calculations
• Data-Enhanced Variational Monte Carlo Simulations for Rydberg Atom Arrays
• Correlation-Enhanced Neural Networks as Interpretable Variational Quantum States
• Dynamical large deviations of two-dimensional kinetically constrained models using a neural-network state ansatz
• Neural Network Evolution Strategy for Solving Quantum Sign Structures
• Simulating disordered quantum systems via dense and sparse restricted Boltzmann machines
• lattice-symmetries: A package for working with quantum many-body bases
• Certification of quantum states with hidden structure of their bitstrings
• jVMC: Versatile and performant variational Monte Carlo leveraging automated differentiation and GPU acceleration
• Gaussian Process States: A data-driven representation of quantum many-body physics
• Enhancing variational Monte Carlo using a programmable quantum simulator
• Reconstruction of classical skyrmions from Anderson towers: quantum Darwinism in action
• Spin-1/2 kagome Heisenberg antiferromagnet: Machine learning discovery of the spinon pair density wave ground state
• Variational optimization of the amplitude of neural-network quantum many-body ground states
• Enhancing quantum state tomography via resource-efficient attention-based neural networks
• The quantum Gaussian process state: A kernel-inspired state with quantum support data
• Learning ground states of gapped quantum Hamiltonians with Kernel Methods
• On the generalizability of artificial neural networks in spin models
• A framework for efficient ab initio electronic structure with Gaussian Process States
• A Bayesian Inference Framework for Compression and Prediction of Quantum States
• Compact Neural-network Quantum State representations of Jastrow and Stabilizer states
• Efficient neural-network based variational Monte Carlo scheme for direct optimization of excited energy states in frustrated quantum systems
• Impact of conditional modelling for a universal autoregressive quantum state
• Probing Criticality in Quantum Spin Chains with Neural Networks
• Many-body quantum sign structures as non-glassy Ising models
• Infinite Neural Network Quantum States: Entanglement and Training Dynamics
• Hybrid Ground-State Quantum Algorithms based on Neural Schrödinger Forging
• Neural-network Quantum States for Spin-1 systems: spin-basis and parameterization effects on compactness of representations
• Rayleigh-Gauss-Newton optimization with enhanced sampling for variational Monte Carlo
• Lee-Yang theory of quantum phase transitions with neural network quantum states
• Investigating Network Parameters in Neural-Network Quantum States
• 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
• Noncentrosymmetric Triangular Magnet CaMnTeO${}_6$: Strong Quantum Fluctuations and Role of s0 vs. s2 Electronic States in Competing Exchange Interactions
• Boltzmann machines and quantum many-body problems
• Broken-Symmetry Ground States of the Heisenberg model on the Pyrochlore Lattice
• Neural-network quantum state study of the long-range antiferromagnetic Ising chain
• Achieving the volume-law entropy regime with random-sign Dicke states
• Quantum Machine Learning of Molecular Energies with Hybrid Quantum-Neural Wavefunction
• Sampling Problems on a Quantum Computer
• Efficiency of the hidden fermion determinant states Ansatz in the light of different complexity measures
• Understanding and eliminating spurious modes in variational Monte Carlo using collective variables
• Improving neural network performance for solving quantum sign structure
• Exploring entanglement in finite-size quantum systems with degenerate ground state
• Phase probabilities in first-order transitions using machine learning
• Discovering quasiorder parameters in the Potts model: A bridge between machine learning and critical phenomena
• Many-Body Neural Network Wavefunction for a Non-Hermitian Ising Chain
• Bound on entanglement in neural quantum states
• Emergence of global receptive fields capturing multipartite quantum correlations
• Universal Performance Gap of Neural Quantum States Applied to the Hofstadter-Bose-Hubbard Model
• Kinetic samplers for neural quantum states
• Bridging the Gap between Deep Learning and Frustrated Quantum Spin System for Extreme-scale Simulations on New Generation of Sunway Supercomputer
• Optimizing Temperature Distributions for Training Neural Quantum States using Parallel Tempering
• Brute-force positivization of $J_1-J_2$ model ground states
• Generalized Lanczos method for systematic optimization of neural-network quantum states
• Principle of learning sign rules by neural networks in qubit lattice models