(Non-) asymptotic properties of Stochastic Gradient Langevin Dynamics

Abstract

Applying standard Markov chain Monte Carlo (MCMC) algorithms to large data sets is computationally infeasible. The recently proposed stochastic gradient Langevin dynamics (SGLD) method circumvents this problem in three ways: it generates proposed moves using only a subset of the data, it skips the Metropolis-Hastings accept-reject step, and it uses sequences of decreasing step sizes. In \cite{TehThierryVollmerSGLD2014}, we provided the mathematical foundations for the decreasing step size SGLD, including consistency and a central limit theorem. However, in practice the SGLD is run for a relatively small number of iterations, and its step size is not decreased to zero. The present article investigates the behaviour of the SGLD with fixed step size. In particular we characterise the asymptotic bias explicitly, along with its dependence on the step size and the variance of the stochastic gradient. On that basis a modified SGLD which removes the asymptotic bias due to the variance of the stochastic gradients up to first order in the step size is derived. Moreover, we are able to obtain bounds on the finite-time bias, variance and mean squared error (MSE). The theory is illustrated with a Gaussian toy model for which the bias and the MSE for the estimation of moments can be obtained explicitly. For this toy model we study the gain of the SGLD over the standard Euler method in the limit of large data sets.

42 pages, 7 figures

References in corpus (1)

• Stochastic Gradient Hamiltonian Monte Carlo

Cited by in corpus (14)

• Three Factors Influencing Minima in SGD
• On the Convergence of Stochastic Gradient MCMC Algorithms with High-Order Integrators
• Cyclical Stochastic Gradient MCMC for Bayesian Deep Learning
• Privacy for Free: Posterior Sampling and Stochastic Gradient Monte Carlo
• Bridging the Gap between Stochastic Gradient MCMC and Stochastic Optimization
• The Fundamental Incompatibility of Hamiltonian Monte Carlo and Data Subsampling
• Practical bounds on the error of Bayesian posterior approximations: A nonasymptotic approach
• Covariance-Controlled Adaptive Langevin Thermostat for Large-Scale Bayesian Sampling
• Enabling scalable stochastic gradient-based inference for Gaussian processes by employing the Unbiased LInear System SolvEr (ULISSE)
• Validated Variational Inference via Practical Posterior Error Bounds
• Scalable Natural Gradient Langevin Dynamics in Practice
• Quantifying the accuracy of approximate diffusions and Markov chains
• Multilevel Monte Carlo for Scalable Bayesian Computations
• Stochastic Gradient MCMC with Stale Gradients