Investigating Network Parameters in Neural-Network Quantum States

TL;DR

This paper explores how neural network parameters in restricted Boltzmann machines reflect quantum phase transitions in the 1D transverse-field Ising model, offering insights into the physical significance of learned parameters.

Contribution

It demonstrates a direct link between RBM parameters and quantum phases, providing a new method to interpret neural-network quantum states.

Findings

RBM parameters reflect quantum phase transition behavior

Finite-temperature phase diagram of RBM correlates with quantum phases

Neural-network parameters can reveal physical insights into quantum states

Abstract

Recently, quantum-state representation using artificial neural networks has started to be recognized as a powerful tool. However, due to the black-box nature of machine learning, it is difficult to analyze what machine learns or why it is powerful. Here, by applying one of the simplest neural networks, the restricted Boltzmann machine (RBM), to the ground-state representation of the one-dimensional (1D) transverse-field Ising (TFI) model, we make an attempt to directly analyze the optimized network parameters. In the RBM optimization, a zero-temperature quantum state is mapped onto a finite-temperature classical state of the extended Ising spins that constitute the RBM. We find that the quantum phase transition from the ordered phase to the disordered phase in the 1D TFI model by increasing the transverse field is clearly reflected in the behaviors of the optimized RBM parameters and hence in the finite-temperature phase diagram of the classical RBM Ising system. The present finding of a correspondence between the neural-network parameters and quantum phases suggests that a careful investigation of the neural-network parameters may provide a new route to extracting nontrivial physical insights from the neural-network wave functions.