Matches in SemOpenAlex for { <https://semopenalex.org/work/W4365440904> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4365440904 abstract "Passive observational data, such as human videos, is abundant and rich in information, yet remains largely untapped by current RL methods. Perhaps surprisingly, we show that passive data, despite not having reward or action labels, can still be used to learn features that accelerate downstream RL. Our approach learns from passive data by modeling intentions: measuring how the likelihood of future outcomes change when the agent acts to achieve a particular task. We propose a temporal difference learning objective to learn about intentions, resulting in an algorithm similar to conventional RL, but which learns entirely from passive data. When optimizing this objective, our agent simultaneously learns representations of states, of policies, and of possible outcomes in an environment, all from raw observational data. Both theoretically and empirically, this scheme learns features amenable for value prediction for downstream tasks, and our experiments demonstrate the ability to learn from many forms of passive data, including cross-embodiment video data and YouTube videos." @default.
- W4365440904 created "2023-04-15" @default.
- W4365440904 creator A5026322200 @default.
- W4365440904 creator A5052979358 @default.
- W4365440904 creator A5056159807 @default.
- W4365440904 date "2023-04-10" @default.
- W4365440904 modified "2023-09-28" @default.
- W4365440904 title "Reinforcement Learning from Passive Data via Latent Intentions" @default.
- W4365440904 doi "https://doi.org/10.48550/arxiv.2304.04782" @default.
- W4365440904 hasPublicationYear "2023" @default.
- W4365440904 type Work @default.
- W4365440904 citedByCount "0" @default.
- W4365440904 crossrefType "posted-content" @default.
- W4365440904 hasAuthorship W4365440904A5026322200 @default.
- W4365440904 hasAuthorship W4365440904A5052979358 @default.
- W4365440904 hasAuthorship W4365440904A5056159807 @default.
- W4365440904 hasBestOaLocation W43654409041 @default.
- W4365440904 hasConcept C102483320 @default.
- W4365440904 hasConcept C105795698 @default.
- W4365440904 hasConcept C119857082 @default.
- W4365440904 hasConcept C121332964 @default.
- W4365440904 hasConcept C127413603 @default.
- W4365440904 hasConcept C132964779 @default.
- W4365440904 hasConcept C145420912 @default.
- W4365440904 hasConcept C154945302 @default.
- W4365440904 hasConcept C15744967 @default.
- W4365440904 hasConcept C199360897 @default.
- W4365440904 hasConcept C201995342 @default.
- W4365440904 hasConcept C21547014 @default.
- W4365440904 hasConcept C23131810 @default.
- W4365440904 hasConcept C2776207758 @default.
- W4365440904 hasConcept C2780451532 @default.
- W4365440904 hasConcept C2780791683 @default.
- W4365440904 hasConcept C33923547 @default.
- W4365440904 hasConcept C37228920 @default.
- W4365440904 hasConcept C41008148 @default.
- W4365440904 hasConcept C62520636 @default.
- W4365440904 hasConcept C97541855 @default.
- W4365440904 hasConceptScore W4365440904C102483320 @default.
- W4365440904 hasConceptScore W4365440904C105795698 @default.
- W4365440904 hasConceptScore W4365440904C119857082 @default.
- W4365440904 hasConceptScore W4365440904C121332964 @default.
- W4365440904 hasConceptScore W4365440904C127413603 @default.
- W4365440904 hasConceptScore W4365440904C132964779 @default.
- W4365440904 hasConceptScore W4365440904C145420912 @default.
- W4365440904 hasConceptScore W4365440904C154945302 @default.
- W4365440904 hasConceptScore W4365440904C15744967 @default.
- W4365440904 hasConceptScore W4365440904C199360897 @default.
- W4365440904 hasConceptScore W4365440904C201995342 @default.
- W4365440904 hasConceptScore W4365440904C21547014 @default.
- W4365440904 hasConceptScore W4365440904C23131810 @default.
- W4365440904 hasConceptScore W4365440904C2776207758 @default.
- W4365440904 hasConceptScore W4365440904C2780451532 @default.
- W4365440904 hasConceptScore W4365440904C2780791683 @default.
- W4365440904 hasConceptScore W4365440904C33923547 @default.
- W4365440904 hasConceptScore W4365440904C37228920 @default.
- W4365440904 hasConceptScore W4365440904C41008148 @default.
- W4365440904 hasConceptScore W4365440904C62520636 @default.
- W4365440904 hasConceptScore W4365440904C97541855 @default.
- W4365440904 hasLocation W43654409041 @default.
- W4365440904 hasOpenAccess W4365440904 @default.
- W4365440904 hasPrimaryLocation W43654409041 @default.
- W4365440904 hasRelatedWork W2063403856 @default.
- W4365440904 hasRelatedWork W2081614221 @default.
- W4365440904 hasRelatedWork W2416943787 @default.
- W4365440904 hasRelatedWork W2979300045 @default.
- W4365440904 hasRelatedWork W3022038857 @default.
- W4365440904 hasRelatedWork W3132818773 @default.
- W4365440904 hasRelatedWork W4246751904 @default.
- W4365440904 hasRelatedWork W4297725807 @default.
- W4365440904 hasRelatedWork W4319083788 @default.
- W4365440904 hasRelatedWork W4319994054 @default.
- W4365440904 isParatext "false" @default.
- W4365440904 isRetracted "false" @default.
- W4365440904 workType "article" @default.