Boosting Monte Carlo simulations of spin glasses using autoregressive neural networks

Abstract

The autoregressive neural networks are emerging as a powerful computational tool to solve relevant problems in classical and quantum mechanics. One of their appealing functionalities is that, after they have learned a probability distribution from a dataset, they allow exact and efficient sampling of typical system configurations. Here we employ a neural autoregressive distribution estimator (NADE) to boost Markov chain Monte Carlo (MCMC) simulations of a paradigmatic classical model of spin-glass theory, namely the two-dimensional Edwards-Anderson Hamiltonian. We show that a NADE can be trained to accurately mimic the Boltzmann distribution using unsupervised learning from system configurations generated using standard MCMC algorithms. The trained NADE is then employed as smart proposal distribution for the Metropolis-Hastings algorithm. This allows us to perform efficient MCMC simulations, which provide unbiased results even if the expectation value corresponding to the probability distribution learned by the NADE is not exact. Notably, we implement a sequential tempering procedure, whereby a NADE trained at a higher temperature is iteratively employed as proposal distribution in a MCMC simulation run at a slightly lower temperature. This allows one to efficiently simulate the spin-glass model even in the low-temperature regime, avoiding the divergent correlation times that plague MCMC simulations driven by local-update algorithms. Furthermore, we show that the NADE-driven simulations quickly sample ground-state configurations, paving the way to their future utilization to tackle binary optimization problems.

arXiv comment13 pages, 14 figures

References in corpus (37)

• Auto-Encoding Variational Bayes
• Exchange Monte Carlo Method and Application to Spin Glass Simulations
• Machine learning and the physical sciences
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Pixel Recurrent Neural Networks
• Quantum annealing with more than one hundred qubits
• Many-body quantum state tomography with neural networks
• Worm algorithms for classical statistical models
• Reconstructing quantum states with generative models
• Neural-Network Approach to Dissipative Quantum Many-Body Dynamics
• Recurrent Neural Network Wave Functions
• Deep autoregressive models for the efficient variational simulation of many-body quantum systems
• Learning Thermodynamics with Boltzmann Machines
• Self-Learning Monte Carlo Method
• Accelerate Monte Carlo Simulations with Restricted Boltzmann Machines
• 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
• Solving Statistical Mechanics Using Variational Autoregressive Networks
• Replica Symmetry Breaking in Short-Range Spin Glasses: Theoretical Foundations and Numerical Evidences
• A Neural Decoder for Topological Codes
• Constructing neural stationary states for open quantum many-body systems
• A Cluster Monte Carlo Algorithm for 2-Dimensional Spin Glasses
• Lower Critical Dimension of Ising Spin Glasses
• Asymptotically unbiased estimation of physical observables with neural samplers
• Efficient Cluster Algorithm for Spin Glasses in Any Space Dimension
• The critical exponents of the two-dimensional Ising spin glass revisited: Exact Ground State Calculations and Monte Carlo Simulations
• Finite Size Scaling Analysis of Exact Ground States for +/-J Spin Glass Models in Two Dimensions
• Aspect-Ratio Scaling and The Stiffness Exponent $\theta$ for Ising Spin Glasses
• Learning hard quantum distributions with variational autoencoders
• Evidence against a mean field description of short-range spin glasses revealed through thermal boundary conditions
• Variational autoencoder reconstruction of complex many-body physics
• Self-learning projective quantum Monte Carlo simulations guided by restricted Boltzmann machines
• Number of thermodynamic states in the three-dimensional Edwards-Anderson spin glass
• Computational hardness of spin-glass problems with tile-planted solutions
• Ultrametricity in 3D Edwards-Anderson spin glasses
• Patch-planting spin-glass solution for benchmarking
• Towards an optimal flow: Density-of-states-informed replica-exchange simulations

Cited by in corpus (32)

• Machine Learning for Quantum Matter
• Unbiased Monte Carlo Cluster Updates with Autoregressive Neural Networks
• Roadmap on machine learning glassy dynamics
• Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems
• Scalable neural networks for the efficient learning of disordered quantum systems
• Power-grid stability predictions using transferable machine learning
• A Bayesian generative neural network framework for epidemic inference problems
• Analysis of autocorrelation times in Neural Markov Chain Monte Carlo simulations
• Cluster Percolation in the Two-Dimensional Ising Spin Glass
• Neural networks in quantum many-body physics: a hands-on tutorial
• Hierarchical autoregressive neural networks for statistical systems
• Normalizing flows as an enhanced sampling method for atomistic supercooled liquids
• Many-Qudit representation for the Travelling Salesman Problem Optimisation
• The autoregressive neural network architecture of the Boltzmann distribution of pairwise interacting spins systems
• Accelerating equilibrium spin-glass simulations using quantum annealers via generative deep learning
• Probing Criticality in Quantum Spin Chains with Neural Networks
• Creating equilibrium glassy states via random particle bonding
• Generative Neural Samplers for the Quantum Heisenberg Chain
• Message Passing Variational Autoregressive Network for Solving Intractable Ising Models
• Network-Initialized Monte Carlo Based on Generative Neural Networks
• Sparse Autoregressive Neural Networks for Classical Spin Systems
• Adaptive Monte Carlo augmented with normalizing flows
• Performance of machine-learning-assisted Monte Carlo in sampling from simple statistical physics models
• Fluctuation based interpretable analysis scheme for quantum many-body snapshots
• Efficient Optimization of Variational Autoregressive Networks with Natural Gradient
• Reducing rejection exponentially improves Markov chain Monte Carlo sampling
• Cluster percolation in the three-dimensional $\pm J$ random-bond Ising model
• Discovering quasiorder parameters in the Potts model: A bridge between machine learning and critical phenomena
• A method for quantifying the generalization capabilities of generative models for solving Ising models
• Nearest-Neighbours Neural Network architecture for efficient sampling of statistical physics models
• Neural-network-assisted Monte Carlo sampling trained by Quantum Approximate Optimization Algorithm
• Ratio Divergence Learning Using Target Energy in Restricted Boltzmann Machines: Beyond Kullback--Leibler Divergence Learning