FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2617832762> ?p ?o ?g. }

• W2617832762 abstract "Typical reinforcement learning (RL) agents learn to complete tasks specified by reward functions tailored to their domain. As such, the policies they learn do not generalize even to similar domains. To address this issue, we develop a framework through which a deep RL agent learns to generalize policies from smaller, simpler domains to more complex ones using a recurrent attention mechanism. The task is presented to the agent as an image and an instruction specifying the goal. This meta-controller guides the agent towards its goal by designing a sequence of smaller subtasks on the part of the state space within the attention, effectively decomposing it. As a baseline, we consider a setup without attention as well. Our experiments show that the meta-controller learns to create subgoals within the attention." @default.
• W2617832762 created "2017-06-05" @default.
• W2617832762 creator A5000008086 @default.
• W2617832762 creator A5009018152 @default.
• W2617832762 creator A5058921365 @default.
• W2617832762 creator A5084125996 @default.
• W2617832762 creator A5091212723 @default.
• W2617832762 date "2017-05-24" @default.
• W2617832762 modified "2023-09-23" @default.
• W2617832762 title "State Space Decomposition and Subgoal Creation for Transfer in Deep Reinforcement Learning" @default.
• W2617832762 cites W1515851193 @default.
• W2617832762 cites W2064675550 @default.
• W2617832762 cites W2119717200 @default.
• W2617832762 cites W2335959470 @default.
• W2617832762 cites W567721252 @default.
• W2617832762 hasPublicationYear "2017" @default.
• W2617832762 type Work @default.
• W2617832762 sameAs 2617832762 @default.
• W2617832762 citedByCount "0" @default.
• W2617832762 crossrefType "posted-content" @default.
• W2617832762 hasAuthorship W2617832762A5000008086 @default.
• W2617832762 hasAuthorship W2617832762A5009018152 @default.
• W2617832762 hasAuthorship W2617832762A5058921365 @default.
• W2617832762 hasAuthorship W2617832762A5084125996 @default.
• W2617832762 hasAuthorship W2617832762A5091212723 @default.
• W2617832762 hasConcept C105795698 @default.
• W2617832762 hasConcept C111368507 @default.
• W2617832762 hasConcept C111919701 @default.
• W2617832762 hasConcept C11413529 @default.
• W2617832762 hasConcept C119857082 @default.
• W2617832762 hasConcept C124681953 @default.
• W2617832762 hasConcept C12725497 @default.
• W2617832762 hasConcept C127313418 @default.
• W2617832762 hasConcept C127413603 @default.
• W2617832762 hasConcept C134306372 @default.
• W2617832762 hasConcept C154945302 @default.
• W2617832762 hasConcept C18903297 @default.
• W2617832762 hasConcept C201995342 @default.
• W2617832762 hasConcept C203479927 @default.
• W2617832762 hasConcept C2778112365 @default.
• W2617832762 hasConcept C2778572836 @default.
• W2617832762 hasConcept C2780451532 @default.
• W2617832762 hasConcept C33923547 @default.
• W2617832762 hasConcept C36503486 @default.
• W2617832762 hasConcept C41008148 @default.
• W2617832762 hasConcept C48103436 @default.
• W2617832762 hasConcept C54355233 @default.
• W2617832762 hasConcept C6557445 @default.
• W2617832762 hasConcept C72434380 @default.
• W2617832762 hasConcept C86803240 @default.
• W2617832762 hasConcept C97541855 @default.
• W2617832762 hasConceptScore W2617832762C105795698 @default.
• W2617832762 hasConceptScore W2617832762C111368507 @default.
• W2617832762 hasConceptScore W2617832762C111919701 @default.
• W2617832762 hasConceptScore W2617832762C11413529 @default.
• W2617832762 hasConceptScore W2617832762C119857082 @default.
• W2617832762 hasConceptScore W2617832762C124681953 @default.
• W2617832762 hasConceptScore W2617832762C12725497 @default.
• W2617832762 hasConceptScore W2617832762C127313418 @default.
• W2617832762 hasConceptScore W2617832762C127413603 @default.
• W2617832762 hasConceptScore W2617832762C134306372 @default.
• W2617832762 hasConceptScore W2617832762C154945302 @default.
• W2617832762 hasConceptScore W2617832762C18903297 @default.
• W2617832762 hasConceptScore W2617832762C201995342 @default.
• W2617832762 hasConceptScore W2617832762C203479927 @default.
• W2617832762 hasConceptScore W2617832762C2778112365 @default.
• W2617832762 hasConceptScore W2617832762C2778572836 @default.
• W2617832762 hasConceptScore W2617832762C2780451532 @default.
• W2617832762 hasConceptScore W2617832762C33923547 @default.
• W2617832762 hasConceptScore W2617832762C36503486 @default.
• W2617832762 hasConceptScore W2617832762C41008148 @default.
• W2617832762 hasConceptScore W2617832762C48103436 @default.
• W2617832762 hasConceptScore W2617832762C54355233 @default.
• W2617832762 hasConceptScore W2617832762C6557445 @default.
• W2617832762 hasConceptScore W2617832762C72434380 @default.
• W2617832762 hasConceptScore W2617832762C86803240 @default.
• W2617832762 hasConceptScore W2617832762C97541855 @default.
• W2617832762 hasLocation W26178327621 @default.
• W2617832762 hasOpenAccess W2617832762 @default.
• W2617832762 hasPrimaryLocation W26178327621 @default.
• W2617832762 hasRelatedWork W2112401277 @default.
• W2617832762 hasRelatedWork W2145739724 @default.
• W2617832762 hasRelatedWork W2158150115 @default.
• W2617832762 hasRelatedWork W2201750637 @default.
• W2617832762 hasRelatedWork W2344013593 @default.
• W2617832762 hasRelatedWork W2513373085 @default.
• W2617832762 hasRelatedWork W2620290674 @default.
• W2617832762 hasRelatedWork W2741995169 @default.
• W2617832762 hasRelatedWork W2789410350 @default.
• W2617832762 hasRelatedWork W2896783380 @default.
• W2617832762 hasRelatedWork W2970479807 @default.
• W2617832762 hasRelatedWork W2970720334 @default.
• W2617832762 hasRelatedWork W298069310 @default.
• W2617832762 hasRelatedWork W2991156573 @default.
• W2617832762 hasRelatedWork W3009245728 @default.
• W2617832762 hasRelatedWork W3080901109 @default.
• W2617832762 hasRelatedWork W3154976930 @default.
• W2617832762 hasRelatedWork W3173049816 @default.
• W2617832762 hasRelatedWork W3202251623 @default.
• W2617832762 hasRelatedWork W57282082 @default.

• next