REVIEW 7 cited by
AlphaStar Unplugged: Large-Scale Offline Reinforcement Learning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
AlphaStar Unplugged: Large-Scale Offline Reinforcement Learning
read the original abstract
StarCraft II is one of the most challenging simulated reinforcement learning environments; it is partially observable, stochastic, multi-agent, and mastering StarCraft II requires strategic planning over long time horizons with real-time low-level execution. It also has an active professional competitive scene. StarCraft II is uniquely suited for advancing offline RL algorithms, both because of its challenging nature and because Blizzard has released a massive dataset of millions of StarCraft II games played by human players. This paper leverages that and establishes a benchmark, called AlphaStar Unplugged, introducing unprecedented challenges for offline reinforcement learning. We define a dataset (a subset of Blizzard's release), tools standardizing an API for machine learning methods, and an evaluation protocol. We also present baseline agents, including behavior cloning, offline variants of actor-critic and MuZero. We improve the state of the art of agents using only offline data, and we achieve 90% win rate against previously published AlphaStar behavior cloning agent.
Forward citations
Cited by 7 Pith papers
-
Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learning
Softmax Transformers implement in-context RL through equivalence to weighted softmax TD updates, with error decay under contraction and parameters as global minimizers of pretraining loss.
-
Convergence and Emergence of In-Context Reinforcement Learning with Chain of Thought
With specific linear Transformer parameters, CoT generation equals iterative TD updates, yielding geometric error decay with CoT length until a context-length statistical floor, and those parameters globally minimize ...
-
Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learning
Softmax Transformers with specific parameters implement iterative weighted softmax TD learning for in-context policy evaluation, with evaluation error decaying over layers and those parameters globally minimizing pret...
-
SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior Data
SOPE uses an actor-aligned OPE signal on a held-out validation split to dynamically stop offline stabilization phases in online RL, improving performance up to 45.6% and cutting TFLOPs up to 22x on 25 Minari tasks.
-
Play Like Champions: Counterfactual Feedback Generation in Latent Space
A guided VAE trained on pro StarCraft replays enables four latent-space traversal strategies to produce counterfactual improvement trajectories for amateur players.
-
Parametric Open Source Games
Introduces parametric open-source games as continuous analogues of program equilibria, proves equilibrium existence, and derives an exact coupling threshold for cooperation in symmetric 2x2 games under gradient ascent.
-
SOPE: Stabilizing Off-Policy Evaluation for Online RL with Prior Data
SOPE dynamically controls offline training length in online RL using actor-aligned OPE on validation data to stop when benefits saturate, achieving up to 45.6% better performance and 22x less computation on Minari tasks.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.