SMAC

The 'SMAC' dataset/benchmark is used to evaluate multi-agent reinforcement learning algorithms by providing a set of scenarios that require coordination among agents in competitive and cooperative settings.

Papers using SMAC (15)

CoFlow: Coordinated Few-Step Flow for Offline Multi-Agent Decision Making2026STAIRS-Former: Spatio-Temporal Attention with Interleaved Recursive Structure Transformer for Offline Multi-task Multi-agent Reinforcement Learning2026Centralized Permutation Equivariant Policy for Cooperative Multi-Agent Reinforcement Learning2025Generalizable Agent Modeling for Agent Collaboration-Competition Adaptation with Multi-Retrieval and Dynamic Generation2025HLSMAC: A New Starcraft Multi-agent Challenge For High-level Strategic Decision-making2025Oryx: a Scalable Sequence Model for Many-Agent Coordination in Offline MARL2025Learning Generalizable Skills from Offline Multi-Task Data for Multi-Agent Cooperation2025Stop Wandering, Find the Keys: LLMs Discriminate Key States for Efficient Multi-Agent Exploration2024Is Independent Learning All You Need in the StarCraft Multi-Agent Challenge?2020 · 185 citesMAVEN: Multi-Agent Variational Exploration2019 · 76 citesQTypeMix: Enhancing Multi-Agent Cooperative Strategies through Heterogeneous and Homogeneous Value Decomposition2024 · 8 citesIntrinsic Action Tendency Consistency for Cooperative Multi-Agent Reinforcement Learning2024 · 6 citesPTDE: Personalized Training with Distilled Execution for Multi-Agent Reinforcement Learning2022 · 4 citesBGC: Multi-Agent Group Belief with Graph Clustering2020Efficient Distributed Framework for Collaborative Multi-Agent Reinforcement Learning2022