StarCraft II Multi-Agent Challenge

The 'StarCraft II Multi-Agent Challenge' is a benchmark that contains scenarios for evaluating cooperative Multi-Agent Reinforcement Learning (MARL) algorithms in a complex, real-time strategy game environment.

Papers using StarCraft II Multi-Agent Challenge (7)

SMAC-Talk: A Natural Language Extension of the StarCraft Multi-Agent Challenge for Large Language Models2026Autonomous Partner Selection for Cooperative Multi-Agent Reinforcement Learning2026Multi-Agent Deep Reinforcement Learning Under Constrained Communications2026Transformer World Model for Sample Efficient Multi-Agent Reinforcement Learning2025Hybrid Training for Enhanced Multi-task Generalization in Multi-agent Reinforcement Learning2024Rethinking Individual Global Max in Cooperative Multi-Agent Reinforcement Learning2022 · 8 citesMAC-PO: Multi-Agent Experience Replay via Collective Priority Optimization2023 · 6 cites