Equivalence of restricted Boltzmann machines and tensor network states

Abstract

The restricted Boltzmann machine (RBM) is one of the fundamental building blocks of deep learning. RBM finds wide applications in dimensional reduction, feature extraction, and recommender systems via modeling the probability distributions of a variety of input data including natural images, speech signals, and customer ratings, etc. We build a bridge between RBM and tensor network states (TNS) widely used in quantum many-body physics research. We devise efficient algorithms to translate an RBM into the commonly used TNS. Conversely, we give sufficient and necessary conditions to determine whether a TNS can be transformed into an RBM of given architectures. Revealing these general and constructive connections can cross-fertilize both deep learning and quantum many-body physics. Notably, by exploiting the entanglement entropy bound of TNS, we can rigorously quantify the expressive power of RBM on complex data sets. Insights into TNS and its entanglement capacity can guide the design of more powerful deep learning architectures. On the other hand, RBM can represent quantum many-body states with fewer parameters compared to TNS, which may allow more efficient classical simulations.

arXiv comment18 pages, 12 figures + 2 appendices; Code implementations at https://github.com/yzcj105/rbm2mps

References in corpus (37)

• Fault-tolerant quantum computation by anyons
• The density-matrix renormalization group in the age of matrix product states
• Area laws for the entanglement entropy - a review
• Efficient classical simulation of slightly entangled quantum computations
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• A Practical Introduction to Tensor Networks: Matrix Product States and Projected Entangled Pair States
• Persistent entanglement in arrays of interacting particles
• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• Efficient simulation of one-dimensional quantum many-body systems
• Entanglement renormalization
• Area laws in quantum systems: mutual information and correlations
• Tensor renormalization group approach to 2D classical lattice models
• Criticality, the area law, and the computational power of PEPS
• Classical simulation of quantum many-body systems with a tree tensor network
• Accurate determination of tensor network state of quantum lattice models in two dimensions
• Quantum Entanglement in Neural Network States
• Coarse-graining renormalization by higher-order singular value decomposition
• Why does deep and cheap learning work so well?
• Studying Two Dimensional Systems With the Density Matrix Renormalization Group
• Efficient Representation of Quantum Many-body States with Deep Neural Networks
• Restricted-Boltzmann-Machine Learning for Solving Strongly Correlated Quantum Systems
• Machine Learning Topological States
• PEPS as ground states: degeneracy and topology
• Density Matrix Renormalization Group Method for 2D Classical Models
• Simulation of two-dimensional quantum systems using a tree tensor network that exploits the entropic area law
• Learning Thermodynamics with Boltzmann Machines
• Simulating Strongly Correlated Quantum Systems with Tree Tensor Networks
• Self-Learning Monte Carlo Method
• Accelerate Monte Carlo Simulations with Restricted Boltzmann Machines
• Neural-Network Quantum States, String-Bond States, and Chiral Topological States
• Resonating valence bond states in the PEPS formalism
• Tensor renormalization of quantum many-body systems using projected entangled simplex states
• A topological insight into restricted Boltzmann machines
• A class of highly entangled many-body states that can be efficiently simulated
• Critical properties of homogeneous binary trees
• Efficient evaluation of partition functions of frustrated and inhomogeneous spin systems
• Real-space renormalization yields finite correlations

Cited by in corpus (170)

• Machine learning and the physical sciences
• A high-bias, low-variance introduction to Machine Learning for physicists
• Tensor networks for complex quantum systems
• Reconstructing quantum states with generative models
• The Expressive Power of Parameterized Quantum Circuits
• Neural-Network Approach to Dissipative Quantum Many-Body Dynamics
• Deep autoregressive models for the efficient variational simulation of many-body quantum systems
• Unsupervised Generative Modeling Using Matrix Product States
• Fermionic neural-network states for ab-initio electronic structure
• Study of the Two-Dimensional Frustrated J1-J2 Model with Neural Network Quantum States
• Simulation methods for open quantum many-body systems
• Tensor Networks for Dimensionality Reduction and Large-Scale Optimizations. Part 2 Applications and Future Perspectives
• Backflow Transformations via Neural Networks for Quantum Many-Body Wave-Functions
• Lecture Notes of Tensor Network Contractions
• The Born Supremacy: Quantum Advantage and Training of an Ising Born Machine
• Constructing exact representations of quantum many-body systems with deep neural networks
• Encoding of Matrix Product States into Quantum Circuits of One- and Two-Qubit Gates
• Machine Learning for Quantum Matter
• Integrating Neural Networks with a Quantum Simulator for State Reconstruction
• Machine learning out-of-equilibrium phases of matter
• Dirac-type nodal spin liquid revealed by refined quantum many-body solver using neural-network wave function, correlation ratio, and level spectroscopy
• Tree Tensor Networks for Generative Modeling
• Identifying Quantum Phase Transitions with Adversarial Neural Networks
• Machine learning of phase transitions in the percolation and XY models
• Discriminative Cooperative Networks for Detecting Phase Transitions
• Solving frustrated quantum many-particle models with convolutional neural networks
• Machine Learning by Unitary Tensor Network of Hierarchical Tree Structure
• Restricted Boltzmann Machines for Quantum States with Nonabelian or Anyonic Symmetries
• Quantum Neural Network States: A Brief Review of Methods and Applications
• Information Perspective to Probabilistic Modeling: Boltzmann Machines versus Born Machines
• Neural tensor contractions and the expressive power of deep neural quantum states
• Deep Learning and AdS/CFT
• Neural Gutzwiller-projected variational wave functions
• Chiral Topological Phases from Artificial Neural Networks
• Developments in the Tensor Network -- from Statistical Mechanics to Quantum Entanglement
• Quenches near Ising quantum criticality as a challenge for artificial neural networks
• Machine Learning Many-Body Localization: Search for the Elusive Nonergodic Metal
• An end-to-end trainable hybrid classical-quantum classifier
• Deep Learning the Ising Model Near Criticality
• Self-learning Monte Carlo with Deep Neural Networks
• Modelling Non-Markovian Quantum Processes with Recurrent Neural Networks
• Unifying Neural-network Quantum States and Correlator Product States via Tensor Networks
• Machine Learning Bell Nonlocality in Quantum Many-body Systems
• Neural network representation of tensor network and chiral states
• Helping restricted Boltzmann machines with quantum-state representation by restoring symmetry
• Efficient Representation of Topologically Ordered States with Restricted Boltzmann Machines
• Experimental quantum homodyne tomography via machine learning
• Geometry of learning neural quantum states
• TensorNetwork for Machine Learning
• Supervised Learning with Projected Entangled Pair States
• Machine Learning meets Quantum Foundations: A Brief Survey
• Self-Learning Monte Carlo Method: Continuous-Time Algorithm
• Smallest Neural Network to Learn the Ising Criticality
• Quantum dynamics in transverse-field Ising models from classical networks
• The learnability scaling of quantum states: restricted Boltzmann machines
• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Tropical Tensor Network for Ground States of Spin Glasses
• Matrix Product Operators for Sequence to Sequence Learning
• Neural-network states for the classical simulation of quantum computing
• Method to solve quantum few-body problems with artificial neural networks
• Interpretable machine learning for inferring the phase boundaries in a nonequilibrium system
• Neural-network quantum states for many-body physics
• Tensor networks for quantum machine learning
• Projective quantum Monte Carlo simulations guided by unrestricted neural network states
• Generalized Transfer Matrix States from Artificial Neural Networks
• Equilibrium and non-Equilibrium regimes in the learning of Restricted Boltzmann Machines
• Deep Learning and Quantum Entanglement: Fundamental Connections with Implications to Network Design
• From Tensor Network Quantum States to Tensorial Recurrent Neural Networks
• Tensor Networks contraction and the Belief Propagation algorithm
• Combined unitary and symmetric group approach applied to low-dimensional spin systems
• Generative machine learning with tensor networks: benchmarks on near-term quantum computers
• QuCumber: wavefunction reconstruction with neural networks
• Quantum neural networks to simulate many-body quantum systems
• Investigating Topological Order using Recurrent Neural Networks
• Efficient Machine Learning Representations of Surface Code with Boundaries, Defects, Domain Walls and Twists
• Extracting Critical Exponent by Finite-Size Scaling with Convolutional Neural Networks
• A perspective on machine learning and data science for strongly correlated electron problems
• Self-learning projective quantum Monte Carlo simulations guided by restricted Boltzmann machines
• Self-learning Monte Carlo method with Behler-Parrinello neural networks
• Neural-network quantum states at finite temperature
• Machine Learning Quantum States -- Extensions to Fermion-Boson Coupled Systems and Excited-State Calculations
• Matrix Product States for Quantum Stochastic Modelling
• Artificial Neural Network Based Computation for Out-of-Time-Ordered Correlators
• Correlation-Enhanced Neural Networks as Interpretable Variational Quantum States
• Approximating power of machine-learning ansatz for quantum many-body states
• Optimal Renormalization Group Transformation from Information Theory
• Identifying Product Order with Restricted Boltzmann Machines
• Opening the black box of deep learning
• On Tensor Train Rank Minimization: Statistical Efficiency and Scalable Algorithm
• Tensor networks for interpretable and efficient quantum-inspired machine learning
• Simulating disordered quantum systems via dense and sparse restricted Boltzmann machines
• Modeling Sequences with Quantum States: A Look Under the Hood
• Adaptive Quantum State Tomography with Active Learning
• Learning the black hole metric from holographic conductivity
• Tensor networks for quantum computing
• Lee-Yang theory of the two-dimensional quantum Ising model
• Restricted Boltzmann Machines and Matrix Product States of 1D Translational Invariant Stabilizer Codes
• Variational Neural-Network Ansatz for Continuum Quantum Field Theory
• Chebyshev expansion of spectral functions using restricted Boltzmann machines
• Machine learning effective models for quantum systems
• Entanglement Features of Random Neural Network Quantum States
• Transforming Generalized Ising Model into Boltzmann Machine
• Atomic Quantum Technologies for Quantum Matter and Fundamental Physics Applications
• Tangent-Space Gradient Optimization of Tensor Network for Machine Learning
• Boltzmann machines as two-dimensional tensor networks
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• Ground state search by local and sequential updates of neural network quantum states
• Quantum State Tomography with Locally Purified Density Operators and Local Measurements
• A Non-stochastic Optimization Algorithm for Neural-network Quantum States
• Legendre Decomposition for Tensors
• Multi-faceted machine learning of competing orders in disordered interacting systems
• Neural-Shadow Quantum State Tomography
• Estimating the Euclidean quantum propagator with deep generative modeling of Feynman paths
• Neural-network quantum states for a two-leg Bose-Hubbard ladder under magnetic flux
• Autoregressive neural Slater-Jastrow ansatz for variational Monte Carlo simulation
• A Bayesian Inference Framework for Compression and Prediction of Quantum States
• Pruning a restricted Boltzmann machine for quantum state reconstruction
• On the descriptive power of Neural-Networks as constrained Tensor Networks with exponentially large bond dimension
• Compact Neural-network Quantum State representations of Jastrow and Stabilizer states
• Sampling scheme for neuromorphic simulation of entangled quantum systems
• Deep learning lattice gauge theories
• An Efficient Algorithmic Way to Construct Boltzmann Machine Representations for Arbitrary Stabilizer Code
• Efficient neural-network based variational Monte Carlo scheme for direct optimization of excited energy states in frustrated quantum systems
• Exponentially improved efficient machine learning for quantum many-body states with provable guarantees
• Measures of distinguishability between stochastic processes
• Tensor Network Computations That Capture Strict Variationality, Volume Law Behavior, and the Efficient Representation of Neural Network States
• Neural-network Quantum States for Spin-1 systems: spin-basis and parameterization effects on compactness of representations
• Invariant Neural Network Ansatz for weakly symmetric Open Quantum Lattices
• Variational Autoencoder Analysis of Ising Model Statistical Distributions and Phase Transitions
• Stack operation of tensor networks
• Neural network enhanced hybrid quantum many-body dynamical distributions
• Entropy, Free Energy, and Work of Restricted Boltzmann Machines
• Two-dimensional local Hamiltonian problem with area laws is QMA-complete
• Classical Quantum Optimization with Neural Network Quantum States
• Residual Matrix Product State for Machine Learning
• Free Energy Evaluation Using Marginalized Annealed Importance Sampling
• Thermodynamics of the Ising model encoded in restricted Boltzmann machines
• Quantum-Classical Machine learning by Hybrid Tensor Networks
• Boltzmann machines and quantum many-body problems
• The resource theory of tensor networks
• Spectroscopy of two-dimensional interacting lattice electrons using symmetry-aware neural backflow transformations
• Exact block encoding of imaginary time evolution with universal quantum neural networks
• Tensor Decompositions in Deep Learning
• Entanglement transitions from restricted Boltzmann machines
• Machine Learning Wavefunction
• The Linear Span of Uniform Matrix Product States
• Non-parametric Semi-Supervised Learning in Many-body Hilbert Space with Rescaled Logarithmic Fidelity
• Variational Neural and Tensor Network Approximations of Thermal States
• Deep tensor networks with matrix product operators
• Finding Quantum Critical Points with Neural-Network Quantum States
• Neural Quantum State Study of Fracton Models
• Neural-network quantum state study of the long-range antiferromagnetic Ising chain
• Classifying topological neural network quantum states via diffusion maps
• Efficiency of the hidden fermion determinant states Ansatz in the light of different complexity measures
• Provably efficient neural network representation for image classification
• Finding the Dynamics of an Integrable Quantum Many-Body System via Machine Learning
• Representing arbitrary ground states of toric code by a restricted Boltzmann machine
• Seeding neural network quantum states with tensor network states
• Representation Theorem for Matrix Product States
• Efficiency of neural-network state representations of one-dimensional quantum spin systems
• Composite Spatial Monte Carlo Integration Based on Generalized Least Squares
• Entanglement Clustering for ground-stateable quantum many-body states
• Neural Tangent Kernel of Matrix Product States: Convergence and Applications
• Basis dependence of Neural Quantum States for the Transverse Field Ising Model
• Quantum Machine Learning for State Tomography Using Classical Data
• Optimized Observable Readout from Single-shot Images of Ultracold Atoms via Machine Learning
• The toric code under antiferromagnetic isotropic Heisenberg interactions
• Matrix Product Operator Restricted Boltzmann Machines
• Learning a compass spin model with neural network quantum states
• Neural-Network Quantum States for Periodic Systems in Continuous Space