Provable Convergence of Variational Monte Carlo Methods

by   Tianyou Li, et al.

The Variational Monte Carlo (VMC) is a promising approach for computing the ground state energy of many-body quantum problems and attracts more and more interests due to the development of machine learning. The recent paradigms in VMC construct neural networks as trial wave functions, sample quantum configurations using Markov chain Monte Carlo (MCMC) and train neural networks with stochastic gradient descent (SGD) method. However, the theoretical convergence of VMC is still unknown when SGD interacts with MCMC sampling given a well-designed trial wave function. Since MCMC reduces the difficulty of estimating gradients, it has inevitable bias in practice. Moreover, the local energy may be unbounded, which makes it harder to analyze the error of MCMC sampling. Therefore, we assume that the local energy is sub-exponential and use the Bernstein inequality for non-stationary Markov chains to derive error bounds of the MCMC estimator. Consequently, VMC is proven to have a first order convergence rate O(log K/√(n K)) with K iterations and a sample size n. It partially explains how MCMC influences the behavior of SGD. Furthermore, we verify the so-called correlated negative curvature condition and relate it to the zero-variance phenomena in solving eigenvalue functions. It is shown that VMC escapes from saddle points and reaches (ϵ,ϵ^1/4) -approximate second order stationary points or ϵ^1/2-variance points in at least O(ϵ^-11/2log^2(1/ϵ) ) steps with high probability. Our analysis enriches the understanding of how VMC converges efficiently and can be applied to general variational methods in physics and statistics.


page 1

page 2

page 3

page 4


A Convergence Analysis for A Class of Practical Variance-Reduction Stochastic Gradient MCMC

Stochastic gradient Markov Chain Monte Carlo (SG-MCMC) has been develope...

Overcoming barriers to scalability in variational quantum Monte Carlo

The variational quantum Monte Carlo (VQMC) method received significant a...

Deep autoregressive models for the efficient variational simulation of many-body quantum systems

Artificial Neural Networks were recently shown to be an efficient repres...

Globally-centered autocovariances in MCMC

Autocovariances are a fundamental quantity of interest in Markov chain M...

SGD with Variance Reduction beyond Empirical Risk Minimization

We introduce a doubly stochastic proximal gradient algorithm for optimiz...

Towards Sampling from Nondirected Probabilistic Graphical models using a D-Wave Quantum Annealer

A D-Wave quantum annealer (QA) having a 2048 qubit lattice, with no miss...

Rayleigh-Gauss-Newton optimization with enhanced sampling for variational Monte Carlo

Variational Monte Carlo (VMC) is an approach for computing ground-state ...

Please sign up or login with your details

Forgot password? Click here to reset