Multiple Mean-Payoff Optimization Under Local Stability Constraints

Authors

  • David Klaška Masaryk University
  • Antonín Kučera Masaryk University
  • Vojtěch Kůr Masaryk University
  • Vít Musil Masaryk University
  • Vojtěch Řehák Masaryk University

DOI:

https://doi.org/10.1609/aaai.v39i25.34856

Abstract

The long-run average payoff per transition (mean payoff) is the main tool for specifying the performance and dependability properties of discrete systems. The problem of constructing a controller (strategy) simultaneously optimizing several mean payoffs has been deeply studied for stochastic and game-theoretic models. One common issue of the constructed controllers is the instability of the mean payoffs, measured by the deviations of the average rewards per transition computed in a finite "window" sliding along a run. Unfortunately, the problem of simultaneously optimizing the mean payoffs under local stability constraints is computationally hard, and the existing works do not provide a practically usable algorithm even for non-stochastic models such as two-player games. In this paper, we design and evaluate the first efficient and scalable solution to this problem applicable to Markov decision processes.

Published

2025-04-11

How to Cite

Klaška, D., Kučera, A., Kůr, V., Musil, V., & Řehák, V. (2025). Multiple Mean-Payoff Optimization Under Local Stability Constraints. Proceedings of the AAAI Conference on Artificial Intelligence, 39(25), 26551–26558. https://doi.org/10.1609/aaai.v39i25.34856

Issue

Section

AAAI Technical Track on Planning, Routing, and Scheduling