Quantitative Weak Convergence for Discrete Stochastic Processes

In this paper, we quantitative convergence in $W_2$ for a family of Langevin-like stochastic processes that includes stochastic gradient descent and related gradient-based algorithms. Under certain regularity assumptions, we show that the iterates of these stochastic processes converge to an invariant distribution at a rate of $\tilde{O}\lrp{1/\sqrt{k}}$ where $k$ is the number of steps; this rate is provably tight up to log factors. Our result reduces to a quantitative form of the classical Central Limit Theorem in the special case when the potential is quadratic.

[1]  Thomas A. Courtade,et al.  Existence of Stein Kernels under a Spectral Gap, and Discrepancy Bound , 2017, Annales de l'Institut Henri Poincaré, Probabilités et Statistiques.

[2]  Alain Durmus,et al.  High-dimensional Bayesian inference via the unadjusted Langevin algorithm , 2016, Bernoulli.

[3]  Thomas A. Courtade,et al.  Bounds on the Poincaré constant for convolution measures , 2018, Annales de l'Institut Henri Poincaré, Probabilités et Statistiques.

[4]  Alex Zhai,et al.  The CLT in high dimensions: Quantitative bounds via martingale embedding , 2018, The Annals of Probability.

[5]  Michael I. Jordan,et al.  Sharp Convergence Rates for Langevin Dynamics in the Nonconvex Setting , 2018, ArXiv.

[6]  Qiang Sun,et al.  Statistical Sparse Online Regression: A Diffusion Approximation Perspective , 2018, AISTATS.

[7]  David M. Blei,et al.  Stochastic Gradient Descent as Approximate Bayesian Inference , 2017, J. Mach. Learn. Res..

[8]  Thomas Bonis.  Rates in the Central Limit Theorem and diffusion approximation via Stein's Method , 2015, 1506.06966.

[9]  David M. Blei,et al.  A Variational Analysis of Stochastic Gradient Algorithms , 2016, ICML.

[10]  A. Dalalyan.  Theoretical guarantees for approximate sampling from smooth and log‐concave densities , 2014, 1412.7392.

[11]  Ioannis Chatzigeorgiou,et al.  Bounds on the Lambert Function and Their Application to the Outage Analysis of User Cooperation , 2013, IEEE Communications Letters.

[12]  Sergey G. Bobkov,et al.  Entropic approach to E. Rio’s central limit theorem for W2 transport distance , 2013 .

[13]  E. Rio.  Upper bounds for minimal distances in the central limit theorem , 2009 .

[14]  C. Villani,et al.  Generalization of an Inequality by Talagrand and Links with the Logarithmic Sobolev Inequality , 2000 .

[15]  Boris Polyak,et al.  Acceleration of stochastic approximation by averaging , 1992 .

[16]  D. Ruppert,et al.  Efficient Estimations from a Slowly Convergent Robbins-Monro Process , 1988 .