2000 character limit reached
Stochastic Gradient Hamiltonian Monte Carlo for Non-Convex Learning (1903.10328v3)
Published 25 Mar 2019 in stat.ML and cs.LG
Abstract: Stochastic Gradient Hamiltonian Monte Carlo (SGHMC) is a momentum version of stochastic gradient descent with properly injected Gaussian noise to find a global minimum. In this paper, non-asymptotic convergence analysis of SGHMC is given in the context of non-convex optimization, where subsampling techniques are used over an i.i.d dataset for gradient updates. Our results complement those of [RRT17] and improve on those of [GGZ18].