Xiao-Qi Sun,Tamra Nebabu,Xizhi Han,Michael O. Flynn,Xiao-Liang Qi
Abstract:Restricted Boltzmann machines (RBMs) are a class of neural networks that have been successfully employed as a variational ansatz for quantum many-body wave functions. Here, we develop an analytic method to study quantum many-body spin states encoded by random RBMs with independent and identically distributed complex Gaussian weights. By mapping the computation of ensemble-averaged quantities to statistical mechanics models, we are able to investigate the parameter space of the RBM ensemble in the thermodynamic limit. We discover qualitatively distinct wave functions by varying RBM parameters, which correspond to distinct phases in the equivalent statistical mechanics model. Notably, there is a regime in which the typical RBM states have near-maximal entanglement entropy in the thermodynamic limit, similar to that of Haar-random states. However, these states generically exhibit nonergodic behavior in the Ising basis, and do not form quantum state designs, making them distinguishable from Haar-random states.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is the effective description of wave functions in quantum many - body systems, especially in how to use Random Restricted Boltzmann Machines (RBMs) to parameterize these wave functions efficiently. Specifically, the paper explores the problem of studying quantum many - body spin states represented by random RBMs encoded by independent and identically distributed complex Gaussian weights through analytical methods. By mapping the calculation of ensemble - averaged quantities onto statistical mechanics models, the authors are able to study the parameter space of RBMs ensembles in the thermodynamic limit. They discovered different wave - function phases, which correspond to different phases in equivalent statistical mechanics models. In particular, they found a region where typical RBM states have entanglement entropy close to the maximum in the thermodynamic limit, similar to Haar - random states. However, these states usually exhibit non - ergodic behavior in the Ising basis and do not form quantum state designs, which distinguishes them from Haar - random states.
The core objective of the paper is to characterize many - body quantum spin states encoded by single - layer random Gaussian RBMs through analytical means, with finite - size numerical simulations supporting the conclusions. Using techniques similar to those in random tensor network research, the authors are able to describe the ensemble - averaged properties of RBM states, such as the statistical fluctuations of their second Rényi entropy and variational wave - function norms. In particular, they found that the calculation of these quantities can be mapped onto classical statistical mechanics models of (coupled) spin chains of length N. Interestingly, in the thermodynamic limit \(N\rightarrow\infty\), when changing the ensemble parameters, these models will exhibit phase transitions that separate wave functions with qualitatively different properties. Through analytical and numerical methods, the authors showed that within a specific RBM parameter interval, random Gaussian RBM states can exhibit maximum entanglement at the leading order of \(1/N\). However, even within this special interval, the Gaussian RBM ensemble can also be distinguished from the Haar - random state ensemble and does not form a quantum state design.