Stochastic Dimension-reduced Second-order Methods for Policy Optimization
Abstract
In this paper, we propose several new stochastic second-order algorithms for policy optimization that only require gradient and Hessian-vector product in each iteration, making them computationally efficient and comparable to policy gradient methods. Specifically, we propose a dimension-reduced second-order method (DR-SOPO) which repeatedly solves a projected two-dimensional trust region subproblem. We show that DR-SOPO obtains an complexity for reaching approximate first-order stationary condition and certain subspace second-order stationary condition. In addition, we present an enhanced algorithm (DVR-SOPO) which further improves the complexity to based on the variance reduction technique. Preliminary experiments show that our proposed algorithms perform favorably compared with stochastic and variance-reduced policy gradient methods.
Cite
@article{arxiv.2301.12174,
title = {Stochastic Dimension-reduced Second-order Methods for Policy Optimization},
author = {Jinsong Liu and Chenghan Xie and Qi Deng and Dongdong Ge and Yinyu Ye},
journal= {arXiv preprint arXiv:2301.12174},
year = {2023}
}