Neural Quantum States in Mixed Precision

TL;DR

This paper explores the use of mixed-precision arithmetic in neural-network based Variational Monte Carlo methods, demonstrating that sampling can be performed in half precision without losing accuracy, thus improving scalability and efficiency.

Contribution

It provides a theoretical framework and empirical validation for applying mixed-precision arithmetic in VMC, enhancing computational efficiency in quantum many-body simulations.

Findings

Sampling in half precision maintains accuracy in VMC.

Mixed-precision reduces memory and energy consumption.

Theoretical bounds guide mixed-precision application.

Abstract

Scientific computing has long relied on double precision (64-bit floating point) arithmetic to guarantee accuracy in simulations of real-world phenomena. However, the growing availability of hardware accelerators such as Graphics Processing Units (GPUs) has made low-precision formats attractive due to their superior performance, reduced memory footprint, and improved energy efficiency. In this work, we investigate the role of mixed-precision arithmetic in neural-network based Variational Monte Carlo (VMC), a widely used method for solving computationally otherwise intractable quantum many-body systems. We first derive general analytical bounds on the error introduced by reduced precision on Metropolis-Hastings MCMC, and then empirically validate these bounds on the use-case of VMC. We demonstrate that significant portions of the algorithm, in particular, sampling the quantum state, can be executed in half precision without loss of accuracy. More broadly, this work provides a theoretical framework to assess the applicability of mixed-precision arithmetic in machine-learning approaches that rely on MCMC sampling. In the context of VMC, we additionally demonstrate the practical effectiveness of mixed-precision strategies, enabling more scalable and energy-efficient simulations of quantum many-body systems.