Methods › Reinforcement Learning › Video Game Models › AlphaStar

DeepMind AlphaStar

AlphaStar

10 papers tagged archive 2025-07-28

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

AlphaStar is a reinforcement learning agent for tackling the game of Starcraft II. It learns a policy π_θ(aₜ|sₜ, z) = P[aₜ|sₜ, z] using a neural network for parameters θ that receives observations sₜ = (o_(1:t), a_(1:t-1)) 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].

AlphaStar uses numerous types of architecture to incorporate different types of features. Observations of player and enemy units are processed with a Transformer. Scatter connections are used to integrate spatial and non-spatial information. The temporal sequence of observations is processed by a core 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.

The 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(λ) and V-trace are used, as well as a new self-imitation algorithm (UPGO).

Lastly, 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.

Image Credit: Yekun Chai

References
  1. Chai, Yekun. "Deciphering AlphaStar on StarCraft II." (2019). https://cyk1337.github.io/notes/2019/07/21/RL/DRL/Decipher-AlphaStar-on-StarCraft-II/
Code Implementation
  1. https://github.com/opendilab/DI-star
See Code · google-deepmind/alphastar

In Syntology the paper matched by title, arXiv:1902.01724, held, no code (no_code; paper_only). The lookup matched a paper by its title, arXiv:1902.01724, not a method entry. The archive credits no paper with an arXiv id for this method, so whether this answer (about arXiv:1902.01724) is about this method is not checked; it does not say whether this method has code.

Origin paper of this answer, arXiv:1902.01724 (its page on Syntology).

Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.

Papers archive 2025-07-28

archive papers tagged: 10 · with a code link: 7 · where Syntology ran a sample: 3 (2 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (3 of 10 tagged: 2 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument)

Page 1 of 1: papers 1 to 10 of 10, newest first by the archive's date (ties by slug), in archive order.

Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.

Tasks archive 2025-07-28

15 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Starcraft II7
Starcraft6
Reinforcement Learning3
Reinforcement Learning (RL)3
reinforcement-learning3
Decision Making2
AI Agent1
CPU1
Deep Reinforcement Learning1
Diversity1
Imitation Learning1
Language Modelling1
Large Language Model1
Offline RL1
Transfer Learning1

Usage over time archive 2025-07-28

Papers per year tagged with AlphaStar: 2019 to 2023, peak 4 4 0 2019: 1 paper 2019 2020: 1 paper 2020 2021: 4 papers 2021 2022: 1 paper 2022 2023: 3 papers 2023
Papers per year the archive tags with this method, by the paper's archive date (10 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Video Game Models

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections