pith. sign in

arxiv: 1805.09416 · v1 · pith:EYE6V3CAnew · submitted 2018-05-23 · 💻 cs.LG · cs.AI· stat.ML

Adaptive Stochastic Gradient Langevin Dynamics: Taming Convergence and Saddle Point Escape Time

classification 💻 cs.LG cs.AIstat.ML
keywords adaptiveasgldgradientdynamicsiterationslangevinstochasticconverge
0
0 comments X
read the original abstract

In this paper, we propose a new adaptive stochastic gradient Langevin dynamics (ASGLD) algorithmic framework and its two specialized versions, namely adaptive stochastic gradient (ASG) and adaptive gradient Langevin dynamics(AGLD), for non-convex optimization problems. All proposed algorithms can escape from saddle points with at most $O(\log d)$ iterations, which is nearly dimension-free. Further, we show that ASGLD and ASG converge to a local minimum with at most $O(\log d/\epsilon^4)$ iterations. Also, ASGLD with full gradients or ASGLD with a slowly linearly increasing batch size converge to a local minimum with iterations bounded by $O(\log d/\epsilon^2)$, which outperforms existing first-order methods.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.