Abstract:Machine learning, one of today's most rapidly growing interdisciplinary fields, promises an unprecedented perspective for solving intricate quantum many-body problems. Understanding the physical aspects of the representative artificial neural-network states has recently become highly desirable in the applications of machine-learning techniques to quantum many-body physics. In this paper, we explore the data structures that encode the physical features in the network states by studying the quantum entanglement properties, with a focus on the restricted-Boltzmann-machine (RBM) architecture. We prove that the entanglement entropy of all short-range RBM states satisfies an area law for arbitrary dimensions and bipartition geometry. For long-range RBM states, we show by using an exact construction that such states could exhibit volume-law entanglement, implying a notable capability of RBM in representing quantum states with massive entanglement. Strikingly, the neural-network representation for these states is remarkably efficient, in the sense that the number of nonzero parameters scales only linearly with the system size. We further examine the entanglement properties of generic RBM states by randomly sampling the weight parameters of the RBM. We find that their averaged entanglement entropy obeys volume-law scaling, and the meantime strongly deviates from the Page entropy of the completely random pure states. We show that their entanglement spectrum has no universal part associated with random matrix theory and bears a Poisson-type level statistics. Using reinforcement learning, we demonstrate that RBM is capable of finding the ground state (with power-law entanglement) of a model Hamiltonian with a long-range interaction. In addition, we show, through a concrete example of the one-dimensional symmetry-protected topological cluster states, that the RBM representation may also be used as a tool to analytically compute the entanglement spectrum. Our results uncover the unparalleled power of artificial neural networks in representing quantum many-body states regardless of how much entanglement they possess, which paves a novel way to bridge computer-science-based machine-learning techniques to outstanding quantum condensed-matter physics problems.

A random energy approach to deep learning

Hyperparameters Adaptation for Restricted Boltzmann Machines Based on Free Energy

Monotone deep Boltzmann machines

Energy-Based Dropout in Restricted Boltzmann Machines: Why Not Go Random

Beyond Random Matrix Theory for Deep Networks

Deep Learning the Ising Model Near Criticality

Short-sighted deep learning

Deep neural networks from the perspective of ergodic theory

Deep Learning with the Random Neural Network and its Applications

On the Learnability of Deep Random Networks

Quantum Entanglement in Neural Network States

Deep learning systems as complex networks

Statistical signatures of abstraction in deep neural networks

Deep learning and the renormalization group

Deep Ensembles: A Loss Landscape Perspective

Simple Deep Random Model Ensemble

A Deep and Stable Extreme Learning Approach for Classification and Regression

Deep Unsupervised Learning using Nonequilibrium Thermodynamics

Cascade of phase transitions in the training of Energy-based models

Random matrix analysis of deep neural network weight matrices

Nonequilibrium Thermodynamics of Restricted Boltzmann Machines