Policy Iteration for Two-Player General-Sum Stochastic Stackelberg Games

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kudo, Mikoto, Akimoto, Youhei
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912966449823744
author Kudo, Mikoto
Akimoto, Youhei
author_facet Kudo, Mikoto
Akimoto, Youhei
contents We address two-player general-sum stochastic Stackelberg games (SSGs), where the leader's policy is optimized considering the best-response follower whose policy is optimal for its reward under the leader. Existing policy gradient and value iteration approaches for SSGs do not guarantee monotone improvement in the leader's policy under the best-response follower. Consequently, their performance is not guaranteed when their limits are not stationary Stackelberg equilibria (SSEs), which do not necessarily exist. In this paper, we derive a policy improvement theorem for SSGs under the best-response follower and propose a novel policy iteration algorithm that guarantees monotone improvement in the leader's performance. Additionally, we introduce Pareto-optimality as an extended optimality of the SSE and prove that our method converges to the Pareto front when the leader is myopic.
format Preprint
id arxiv_https___arxiv_org_abs_2405_06689
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Policy Iteration for Two-Player General-Sum Stochastic Stackelberg Games
Kudo, Mikoto
Akimoto, Youhei
Computer Science and Game Theory
Machine Learning
Multiagent Systems
Optimization and Control
We address two-player general-sum stochastic Stackelberg games (SSGs), where the leader's policy is optimized considering the best-response follower whose policy is optimal for its reward under the leader. Existing policy gradient and value iteration approaches for SSGs do not guarantee monotone improvement in the leader's policy under the best-response follower. Consequently, their performance is not guaranteed when their limits are not stationary Stackelberg equilibria (SSEs), which do not necessarily exist. In this paper, we derive a policy improvement theorem for SSGs under the best-response follower and propose a novel policy iteration algorithm that guarantees monotone improvement in the leader's performance. Additionally, we introduce Pareto-optimality as an extended optimality of the SSE and prove that our method converges to the Pareto front when the leader is myopic.
title Policy Iteration for Two-Player General-Sum Stochastic Stackelberg Games
topic Computer Science and Game Theory
Machine Learning
Multiagent Systems
Optimization and Control
url https://arxiv.org/abs/2405.06689