The StarCraft Multi-Agent Challenge
In the last few years, deep multi-agent reinforcement learning (RL) has become a highly active area of research. A particularly challenging class of problems in this area is partially observable, cooperative, multi-agent learning, in which teams of agents must learn to coordinate their behaviour whi...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Conference item |
| Published: |
International Foundation for Autonomous Agents and Multiagent Systems
2019
|
| _version_ | 1797067234590326784 |
|---|---|
| author | Mikayel Samvelyan Tabish Rashid Christian Schroeder de Witt Gregory Farquhar Nantas Nardelli Tim GJ Rudner Chia-Man Hung Philip HS Torr Jakob Foerster Shimon Whiteson |
| author_facet | Mikayel Samvelyan Tabish Rashid Christian Schroeder de Witt Gregory Farquhar Nantas Nardelli Tim GJ Rudner Chia-Man Hung Philip HS Torr Jakob Foerster Shimon Whiteson |
| author_sort | Mikayel Samvelyan |
| collection | OXFORD |
| description | In the last few years, deep multi-agent reinforcement learning (RL)
has become a highly active area of research. A particularly challenging class of problems in this area is partially observable, cooperative,
multi-agent learning, in which teams of agents must learn to coordinate their behaviour while conditioning only on their private
observations. This is an attractive research area since such problems are relevant to a large number of real-world systems and are
also more amenable to evaluation than general-sum problems.
Standardised environments such as the ALE and MuJoCo have
allowed single-agent RL to move beyond toy domains, such as grid
worlds. However, there is no comparable benchmark for cooperative multi-agent RL. As a result, most papers in this field use one-off
toy problems, making it difficult to measure real progress. In this
paper, we propose the StarCraft Multi-Agent Challenge (SMAC)
as a benchmark problem to fill this gap.1 SMAC is based on the
popular real-time strategy game StarCraft II and focuses on micromanagement challenges where each unit is controlled by an
independent agent that must act based on local observations. We
offer a diverse set of challenge maps and recommendations for best
practices in benchmarking and evaluations. We also open-source
a deep multi-agent RL learning framework including state-of-theart algorithms.2 We believe that SMAC can provide a standard
benchmark environment for years to come.
Videos of our best agents for several SMAC scenarios are available at: https://youtu.be/VZ7zmQ_obZ0.
|
| first_indexed | 2024-03-06T21:53:28Z |
| format | Conference item |
| id | oxford-uuid:4c1d3084-fc02-4a12-8b2f-6b9f9d43ebd0 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T21:53:28Z |
| publishDate | 2019 |
| publisher | International Foundation for Autonomous Agents and Multiagent Systems |
| record_format | dspace |
| spelling | oxford-uuid:4c1d3084-fc02-4a12-8b2f-6b9f9d43ebd02022-03-26T15:47:31ZThe StarCraft Multi-Agent ChallengeConference itemhttp://purl.org/coar/resource_type/c_5794uuid:4c1d3084-fc02-4a12-8b2f-6b9f9d43ebd0Symplectic ElementsInternational Foundation for Autonomous Agents and Multiagent Systems2019Mikayel SamvelyanTabish RashidChristian Schroeder de WittGregory FarquharNantas NardelliTim GJ RudnerChia-Man HungPhilip HS TorrJakob FoersterShimon WhitesonIn the last few years, deep multi-agent reinforcement learning (RL) has become a highly active area of research. A particularly challenging class of problems in this area is partially observable, cooperative, multi-agent learning, in which teams of agents must learn to coordinate their behaviour while conditioning only on their private observations. This is an attractive research area since such problems are relevant to a large number of real-world systems and are also more amenable to evaluation than general-sum problems. Standardised environments such as the ALE and MuJoCo have allowed single-agent RL to move beyond toy domains, such as grid worlds. However, there is no comparable benchmark for cooperative multi-agent RL. As a result, most papers in this field use one-off toy problems, making it difficult to measure real progress. In this paper, we propose the StarCraft Multi-Agent Challenge (SMAC) as a benchmark problem to fill this gap.1 SMAC is based on the popular real-time strategy game StarCraft II and focuses on micromanagement challenges where each unit is controlled by an independent agent that must act based on local observations. We offer a diverse set of challenge maps and recommendations for best practices in benchmarking and evaluations. We also open-source a deep multi-agent RL learning framework including state-of-theart algorithms.2 We believe that SMAC can provide a standard benchmark environment for years to come. Videos of our best agents for several SMAC scenarios are available at: https://youtu.be/VZ7zmQ_obZ0. |
| spellingShingle | Mikayel Samvelyan Tabish Rashid Christian Schroeder de Witt Gregory Farquhar Nantas Nardelli Tim GJ Rudner Chia-Man Hung Philip HS Torr Jakob Foerster Shimon Whiteson The StarCraft Multi-Agent Challenge |
| title | The StarCraft Multi-Agent Challenge |
| title_full | The StarCraft Multi-Agent Challenge |
| title_fullStr | The StarCraft Multi-Agent Challenge |
| title_full_unstemmed | The StarCraft Multi-Agent Challenge |
| title_short | The StarCraft Multi-Agent Challenge |
| title_sort | starcraft multi agent challenge |
| work_keys_str_mv | AT mikayelsamvelyan thestarcraftmultiagentchallenge AT tabishrashid thestarcraftmultiagentchallenge AT christianschroederdewitt thestarcraftmultiagentchallenge AT gregoryfarquhar thestarcraftmultiagentchallenge AT nantasnardelli thestarcraftmultiagentchallenge AT timgjrudner thestarcraftmultiagentchallenge AT chiamanhung thestarcraftmultiagentchallenge AT philiphstorr thestarcraftmultiagentchallenge AT jakobfoerster thestarcraftmultiagentchallenge AT shimonwhiteson thestarcraftmultiagentchallenge AT mikayelsamvelyan starcraftmultiagentchallenge AT tabishrashid starcraftmultiagentchallenge AT christianschroederdewitt starcraftmultiagentchallenge AT gregoryfarquhar starcraftmultiagentchallenge AT nantasnardelli starcraftmultiagentchallenge AT timgjrudner starcraftmultiagentchallenge AT chiamanhung starcraftmultiagentchallenge AT philiphstorr starcraftmultiagentchallenge AT jakobfoerster starcraftmultiagentchallenge AT shimonwhiteson starcraftmultiagentchallenge |