Shaping Zero-Shot Coordination via State Blocking

TL;DR

This paper introduces State-Blocked Coordination (SBC), a simple framework that enhances zero-shot coordination by creating diverse interaction scenarios through state blocking, leading to better generalization to unseen partners.

Contribution

The paper proposes SBC, a novel method that improves zero-shot coordination by generating virtual environments via state blocking without modifying the environment.

Findings

SBC outperforms existing methods in multiple benchmarks.

SBC achieves strong generalization to human partners.

SBC effectively induces diverse interaction scenarios.

Abstract

Zero-shot coordination (ZSC) aims to enable agents to cooperate with independently trained partners without prior interaction, a key requirement for real-world multi-agent systems and human-AI collaboration. Existing approaches have largely emphasized increasing partner diversity during training, yet such strategies often fall short of achieving reliable generalization to unseen partners. We introduce State-Blocked Coordination (SBC), a simple yet effective framework that improves ZSC by inducing diverse interaction scenarios without direct environment modification. Specifically, SBC generates a family of virtual environments through state blocking, allowing agents to experience a wide range of suboptimal partner policies. Across multiple benchmarks, SBC demonstrates superior performance in zero-shot coordination, including strong generalization to human partners.