{"url":"/method/alphastar","slug":"alphastar","name":"AlphaStar","full_name":"DeepMind AlphaStar","full_name_withheld":false,"description_markdown":"**AlphaStar** is a reinforcement learning agent for tackling the game of Starcraft II. It learns a policy $\\pi\\_{\\theta}\\left(a\\_{t}\\mid{s\\_{t}}, z\\right) = P\\left[a\\_{t}\\mid{s\\_{t}}, z\\right]$ using a neural network for parameters $\\theta$ that receives observations $s\\_{t} = \\left(o\\_{1:t}, a\\_{1:t-1}\\right)$ as inputs and chooses actions as outputs. Additionally, the policy conditions on a statistic $z$ that summarizes a strategy sampled from human data such as a build order [1].\r\n\r\nAlphaStar uses numerous types of architecture to incorporate different types of features. Observations of player and enemy units are processed with a [Transformer](https://paperswithcode.com/method/transformer). Scatter connections are used to integrate spatial and non-spatial information. The temporal sequence of observations is processed by a core [LSTM](https://paperswithcode.com/method/lstm). Minimap features are extracted with a Residual Network. To manage the combinatorial action space, the agent uses an autoregressive policy and a recurrent [pointer network](https://paperswithcode.com/method/pointer-net).\r\n\r\nThe agent is trained first with supervised learning from human replays. Parameters are subsequently trained using reinforcement learning that maximizes the win rate against opponents.  The RL algorithm is based on a policy-gradient algorithm similar to actor-critic. Updates are performed asynchronously and off-policy. To deal with this, a combination of $TD\\left(\\lambda\\right)$ and [V-trace](https://paperswithcode.com/method/v-trace) are used, as well as a new self-imitation algorithm (UPGO).\r\n\r\nLastly, to address game-theoretic challenges, AlphaStar is trained with league training to try to approximate a fictitious self-play (FSP) setting which avoids cycles by computing a best response against a uniform mixture of all previous policies. The league of potential opponents includes a diverse range of agents, including policies from current and previous agents.\r\n\r\nImage Credit: [Yekun Chai](https://cyk1337.github.io/notes/2019/07/21/RL/DRL/Decipher-AlphaStar-on-StarCraft-II/)\r\n\r\n####  References\r\n1. Chai, Yekun. \"Deciphering AlphaStar on StarCraft II.\" (2019).  [https://cyk1337.github.io/notes/2019/07/21/RL/DRL/Decipher-AlphaStar-on-StarCraft-II/](https://cyk1337.github.io/notes/2019/07/21/RL/DRL/Decipher-AlphaStar-on-StarCraft-II/)\r\n\r\n\r\n#### Code Implementation\r\n1. https://github.com/opendilab/DI-star","description_state":"present","introduced_year":null,"introduced_by":{"title":null,"paper":null,"first_author":null,"n_authors":0,"url_abs":null,"archive_paper_url":null},"source":{"url":null,"title":null,"url_on_a_paper_host":false},"code_snippet_url":"https://github.com/google-deepmind/alphastar","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Reinforcement Learning","area_id":"reinforcement-learning","collection":"Video Game Models","url":"/methods/category/video-game-models","pwc_aliases":[]}],"n_papers_tagged":10,"archive_num_papers":10,"papers_newest_first":[{"paper":"/paper/large-language-models-play-starcraft-ii","title":"Large Language Models Play StarCraft II: Benchmarks and A Chain of Summarization Approach","date":"2023-12-19","arxiv_id":"2312.11865","n_code_links":1,"syntology":{"ran":2,"of":2,"unverified":0,"pointer_only":2}},{"paper":null,"title":"A Robust and Opponent-Aware League Training Method for StarCraft II","date":"2023-09-21","arxiv_id":null,"n_code_links":0,"syntology":null},{"paper":"/paper/alphastar-unplugged-large-scale-offline","title":"AlphaStar Unplugged: Large-Scale Offline Reinforcement Learning","date":"2023-08-07","arxiv_id":"2308.03526","n_code_links":1,"syntology":{"ran":4,"of":4,"unverified":0,"pointer_only":0}},{"paper":"/paper/on-efficient-reinforcement-learning-for-full","title":"On Efficient Reinforcement Learning for Full-length Game of StarCraft II","date":"2022-09-23","arxiv_id":"2209.11553","n_code_links":2,"syntology":{"ran":0,"of":10,"unverified":10,"pointer_only":0}},{"paper":null,"title":"AI in Human-computer Gaming: Techniques, Challenges and Opportunities","date":"2021-11-15","arxiv_id":"2111.07631","n_code_links":0,"syntology":null},{"paper":"/paper/rethinking-of-alphastar","title":"Rethinking of AlphaStar","date":"2021-08-07","arxiv_id":"2108.03452","n_code_links":2,"syntology":null},{"paper":"/paper/an-introduction-of-mini-alphastar","title":"An Introduction of mini-AlphaStar","date":"2021-04-14","arxiv_id":"2104.06890","n_code_links":1,"syntology":null},{"paper":"/paper/deep-reinforcement-learning-with-function","title":"Deep Reinforcement Learning with Function Properties in Mean Reversion Strategies","date":"2021-01-09","arxiv_id":"2101.03418","n_code_links":1,"syntology":null},{"paper":"/paper/tstarbot-x-an-open-sourced-and-comprehensive","title":"TStarBot-X: An Open-Sourced and Comprehensive Study for Efficient League Training in StarCraft II Full Game","date":"2020-11-27","arxiv_id":"2011.13729","n_code_links":1,"syntology":null},{"paper":null,"title":"AlphaStar: An Evolutionary Computation Perspective","date":"2019-02-05","arxiv_id":"1902.01724","n_code_links":0,"syntology":null}],"papers_shown":10,"tasks":[{"task":"/task/starcraft-ii","name":"Starcraft II","papers":7},{"task":"/task/starcraft","name":"Starcraft","papers":6},{"task":"/task/reinforcement-learning","name":"Reinforcement Learning","papers":3},{"task":"/task/reinforcement-learning-1","name":"Reinforcement Learning (RL)","papers":3},{"task":"/task/reinforcement-learning-2","name":"reinforcement-learning","papers":3},{"task":"/task/decision-making","name":"Decision Making","papers":2},{"task":"/task/ai-agent","name":"AI Agent","papers":1},{"task":null,"name":"CPU","papers":1},{"task":"/task/deep-reinforcement-learning","name":"Deep Reinforcement Learning","papers":1},{"task":"/task/diversity","name":"Diversity","papers":1},{"task":"/task/imitation-learning","name":"Imitation Learning","papers":1},{"task":"/task/language-modelling","name":"Language Modelling","papers":1},{"task":"/task/large-language-model","name":"Large Language Model","papers":1},{"task":"/task/offline-rl","name":"Offline RL","papers":1},{"task":"/task/transfer-learning","name":"Transfer Learning","papers":1}],"tasks_shown":15,"n_tasks":15,"usage_by_year":[{"year":"2019","papers":1},{"year":"2020","papers":1},{"year":"2021","papers":4},{"year":"2022","papers":1},{"year":"2023","papers":3}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/alphastar"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}