tlooto
Sign up
Computer ScienceMathematics

A Floyd-Warshall approach to value computation in Markov decision processes

Aymeric Côme, Éric Fabre, L. Hélouët

2026.2.1International Journal on Software Tools for Technology Transfer

DOI: 10.1007/s10009-026-00849-x

tlooto Summary

This paper revisits the value and policy iteration paradigm and examines a depth-first search strategy that reformulates the average reward computation as an integral over (future) paths that is better expressed in the formalism of weighted automata.

Abstract

Abstract is not available.

Citation format

CÔME, Aymeric; FABRE, Éric; HÉLOUËT, L. A floyd-warshall approach to value computation in markov decision processes. International Journal on Software Tools for Technology Transfer, 2026, 28(1): 5–25.