Machine learning technique to find quantum many-body ground states of bosons on a lattice

TL;DR

This paper introduces a neural network-based variational approach to find ground states of the Bose-Hubbard model, demonstrating efficiency and versatility in representing different atom numbers.

Contribution

It presents a novel neural network architecture and optimization strategy for accurately approximating many-body ground states in quantum lattice systems.

Findings

Fully-connected single hidden layer networks outperform deeper ones.

Convolutional networks are more efficient than fully-connected networks.

Adaptive gradient methods like AdaGrad and Adam improve optimization.

Abstract

We develop a variational method to obtain many-body ground states of the Bose-Hubbard model using feedforward artificial neural networks. A fully-connected network with a single hidden layer works better than a fully-connected network with multiple hidden layers, and a multi-layer convolutional network is more efficient than a fully-connected network. AdaGrad and Adam are optimization methods that work well. Moreover, we show that many-body ground states with different numbers of atoms can be generated by a single network.