Stable Dual Dynamic Programming

Recently, we have introduced a novel approach to dynamic programming and re- inforcement learning that is based on maintaining explicit representations of sta- tionary distributions instead of value functions. In this paper, we investigate the convergence properties of these dual algorithms both theoretically and empirically, and show how they can be scaled up by incorporating function approximation.

Authors

• Tao Wang
• Michael Bowling University of Alberta
• Dale Schuurmans
• Daniel Lizotte

Keywords

No keywords are indexed for this paper.