Matches in SemOpenAlex for { <https://semopenalex.org/work/W4281663178> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4281663178 abstract "A longstanding goal of the field of AI is a method for learning a highly capable, generalist agent from diverse experience. In the subfields of vision and language, this was largely achieved by scaling up transformer-based models and training them on large, diverse datasets. Motivated by this progress, we investigate whether the same strategy can be used to produce generalist reinforcement learning agents. Specifically, we show that a single transformer-based model - with a single set of weights - trained purely offline can play a suite of up to 46 Atari games simultaneously at close-to-human performance. When trained and evaluated appropriately, we find that the same trends observed in language and vision hold, including scaling of performance with model size and rapid adaptation to new games via fine-tuning. We compare several approaches in this multi-game setting, such as online and offline RL methods and behavioral cloning, and find that our Multi-Game Decision Transformer models offer the best scalability and performance. We release the pre-trained models and code to encourage further research in this direction." @default.
- W4281663178 created "2022-06-13" @default.
- W4281663178 creator A5004083108 @default.
- W4281663178 creator A5014193325 @default.
- W4281663178 creator A5016302450 @default.
- W4281663178 creator A5016594236 @default.
- W4281663178 creator A5019230535 @default.
- W4281663178 creator A5028046389 @default.
- W4281663178 creator A5029652042 @default.
- W4281663178 creator A5030056329 @default.
- W4281663178 creator A5045305703 @default.
- W4281663178 creator A5051461147 @default.
- W4281663178 creator A5057773393 @default.
- W4281663178 date "2022-05-30" @default.
- W4281663178 modified "2023-09-29" @default.
- W4281663178 title "Multi-Game Decision Transformers" @default.
- W4281663178 doi "https://doi.org/10.48550/arxiv.2205.15241" @default.
- W4281663178 hasPublicationYear "2022" @default.
- W4281663178 type Work @default.
- W4281663178 citedByCount "0" @default.
- W4281663178 crossrefType "posted-content" @default.
- W4281663178 hasAuthorship W4281663178A5004083108 @default.
- W4281663178 hasAuthorship W4281663178A5014193325 @default.
- W4281663178 hasAuthorship W4281663178A5016302450 @default.
- W4281663178 hasAuthorship W4281663178A5016594236 @default.
- W4281663178 hasAuthorship W4281663178A5019230535 @default.
- W4281663178 hasAuthorship W4281663178A5028046389 @default.
- W4281663178 hasAuthorship W4281663178A5029652042 @default.
- W4281663178 hasAuthorship W4281663178A5030056329 @default.
- W4281663178 hasAuthorship W4281663178A5045305703 @default.
- W4281663178 hasAuthorship W4281663178A5051461147 @default.
- W4281663178 hasAuthorship W4281663178A5057773393 @default.
- W4281663178 hasBestOaLocation W42816631781 @default.
- W4281663178 hasConcept C119599485 @default.
- W4281663178 hasConcept C119857082 @default.
- W4281663178 hasConcept C127413603 @default.
- W4281663178 hasConcept C154945302 @default.
- W4281663178 hasConcept C165801399 @default.
- W4281663178 hasConcept C166957645 @default.
- W4281663178 hasConcept C41008148 @default.
- W4281663178 hasConcept C48044578 @default.
- W4281663178 hasConcept C66322947 @default.
- W4281663178 hasConcept C77088390 @default.
- W4281663178 hasConcept C79581498 @default.
- W4281663178 hasConcept C95457728 @default.
- W4281663178 hasConcept C97541855 @default.
- W4281663178 hasConceptScore W4281663178C119599485 @default.
- W4281663178 hasConceptScore W4281663178C119857082 @default.
- W4281663178 hasConceptScore W4281663178C127413603 @default.
- W4281663178 hasConceptScore W4281663178C154945302 @default.
- W4281663178 hasConceptScore W4281663178C165801399 @default.
- W4281663178 hasConceptScore W4281663178C166957645 @default.
- W4281663178 hasConceptScore W4281663178C41008148 @default.
- W4281663178 hasConceptScore W4281663178C48044578 @default.
- W4281663178 hasConceptScore W4281663178C66322947 @default.
- W4281663178 hasConceptScore W4281663178C77088390 @default.
- W4281663178 hasConceptScore W4281663178C79581498 @default.
- W4281663178 hasConceptScore W4281663178C95457728 @default.
- W4281663178 hasConceptScore W4281663178C97541855 @default.
- W4281663178 hasLocation W42816631781 @default.
- W4281663178 hasOpenAccess W4281663178 @default.
- W4281663178 hasPrimaryLocation W42816631781 @default.
- W4281663178 hasRelatedWork W1527726406 @default.
- W4281663178 hasRelatedWork W2923653485 @default.
- W4281663178 hasRelatedWork W2957776456 @default.
- W4281663178 hasRelatedWork W2980433389 @default.
- W4281663178 hasRelatedWork W3022038857 @default.
- W4281663178 hasRelatedWork W3035435378 @default.
- W4281663178 hasRelatedWork W3180915449 @default.
- W4281663178 hasRelatedWork W3207201861 @default.
- W4281663178 hasRelatedWork W4287079499 @default.
- W4281663178 hasRelatedWork W4319083788 @default.
- W4281663178 isParatext "false" @default.
- W4281663178 isRetracted "false" @default.
- W4281663178 workType "article" @default.