Paper Image

Efficient solution of factored Markov decision processes

Published on:

2 April 2024

Primary Category:

Optimization and Control

Paper Authors:

Huikang Liu,

Wolfram Wiesemann,

Man-Chung Yue

Bullets

Key Details

Proposes mathematical programming approach to solve FMDPs using off-the-shelf solvers

Exploits factored structure in state and action spaces for efficiency

Extends to robust FMDPs with uncertain transition dynamics

Numerical experiments show potential of approach

AI generated summary

Efficient solution of factored Markov decision processes

This paper proposes a mathematical programming approach to solving factored Markov decision processes (FMDPs). In contrast to existing algorithms, this approach uses off-the-shelf solvers for streamlined implementation and maintenance. It capitalizes on factored structure in state and action spaces and extends to robust FMDPs with uncertain transition dynamics. Numerical experiments demonstrate the potential of this approach.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up