Dynamic programming through the lens of semismooth Newton-type methods

M Gargiani, A Zanelli… - IEEE Control …, 2022 - ieeexplore.ieee.org
Policy iteration and value iteration are at the core of many (approximate) dynamic
programming methods. For Markov Decision Processes with finite state and action spaces …

Dynamic programming through the lens of semismooth Newton-type methods (extended version)

M Gargiani, A Zanelli, D Liao-McPherson… - arxiv preprint arxiv …, 2022 - arxiv.org
Policy iteration and value iteration are at the core of many (approximate) dynamic
programming methods. For Markov Decision Processes with finite state and action spaces …

Inexact GMRES policy iteration for large-scale Markov decision processes

M Gargiani, D Liao-McPherson, A Zanelli, J Lygeros - IFAC-PapersOnLine, 2023 - Elsevier
Policy iteration enjoys a local quadratic rate of contraction, but its iterations are
computationally expensive for Markov decision processes (MDPs) with a large number of …