Learning Optimal Strategies for Temporal Tasks in Stochastic Games

• Published in: IEEE transactions on automatic control 2024-11, Vol.69 (11), p.7387-7402
• Main Authors: Bozkurt, Alper Kamil, Wang, Yu, Zavlanos, Michael M., Pajic, Miroslav
• Format: Article
• Language: English
• Subjects: Algorithms; Color; Controller synthesis; Controllers; Discounts; Game theory; Games; Learning automata; linear temporal logic (LTL); Maximization; Optimization; Reinforcement learning; reinforcement learning (RL); Scalability; Specifications; stochastic games (SGs); Stochastic processes; Synthesis; Task analysis; Temporal logic
• ISSN: 0018-9286; 1558-2523
• DOI: 10.1109/TAC.2024.3390848

Synthesis from linear temporal logic (LTL) specifications provides assured controllers for systems operating in stochastic and potentially adversarial environments. Automatic synthesis tools, however, require a model of the environment to construct controllers. In this work, we introduce a model-free reinforcement learning (RL) approach to derive controllers from given LTL specifications even when the environment is completely unknown. We model the problem as a stochastic game (SG) between the controller and the environment; we then learn optimal strategies that maximize the probability of satisfying the LTL specifications against the worst-case environment behavior. We first construct a product game using the deterministic parity automaton (DPA) translated from the given LTL specification. By deriving distinct rewards and discount factors from the acceptance condition of the DPA, we reduce the maximization of the worst-case probability of satisfying the LTL specification into the maximization of a discounted reward objective in the product game; this enables the use of model-free RL algorithms to learn an optimal controller strategy. To address the scalability issues arising when the number of sets defining the acceptance condition of the DPA, usually referred to as colors, is large; we propose a lazy color generation method where distinct rewards and discount factors are utilized only when needed, and an approximate method where the controller eventually focuses on only one color. In several case studies, we show that our approach is scalable to a wide range of LTL formulas, significantly outperforming existing methods that learn controllers from LTL specifications in SGs.