FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3207570629> ?p ?o ?g. }

• W3207570629 abstract "Abstraction plays an important role for generalisation of knowledge and skills, and is key to sample efficient learning and planning. For many complex problems an abstract plan can be formed first, which is then instantiated by filling in the necessary low-level details. Often, such abstract plans generalize well to related new problems. We study abstraction in the context of reinforcement learning, in which agents may perform state or temporal abstractions. Temporal abstractions aka options represent temporally-extended actions in the form of option policies. However, typically acquired option policies cannot be directly transferred to new environments due to changes in the state space or transition dynamics. Furthermore, many existing state abstraction schemes ignore the correlation between state and temporal abstraction. In this work, we propose successor abstraction, a novel abstraction scheme building on successor features. This includes an algorithm for encoding and instantiation of abstract options across different environments, and a state abstraction mechanism based on the abstract options. Our successor abstraction allows us to learn abstract environment models with semantics that are transferable across different environments through encoding and instantiation of abstract options. Empirically, we achieve better transfer and improved performance on a set of benchmark tasks as compared to relevant state of the art baselines." @default.
• W3207570629 created "2021-10-25" @default.
• W3207570629 creator A5000423552 @default.
• W3207570629 creator A5051794439 @default.
• W3207570629 creator A5065393757 @default.
• W3207570629 date "2021-10-18" @default.
• W3207570629 modified "2023-09-26" @default.
• W3207570629 title "MDP Abstraction with Successor Features." @default.
• W3207570629 cites W1492014007 @default.
• W3207570629 cites W1688218840 @default.
• W3207570629 cites W1964535365 @default.
• W3207570629 cites W1999874108 @default.
• W3207570629 cites W2056354534 @default.
• W3207570629 cites W2061562262 @default.
• W3207570629 cites W2098509290 @default.
• W3207570629 cites W2098774185 @default.
• W3207570629 cites W2101234009 @default.
• W3207570629 cites W2109910161 @default.
• W3207570629 cites W2132057084 @default.
• W3207570629 cites W2145339207 @default.
• W3207570629 cites W2434014514 @default.
• W3207570629 cites W2440926996 @default.
• W3207570629 cites W2468354762 @default.
• W3207570629 cites W2765308067 @default.
• W3207570629 cites W2766447205 @default.
• W3207570629 cites W2809461852 @default.
• W3207570629 cites W2949078258 @default.
• W3207570629 cites W2951548327 @default.
• W3207570629 cites W2965666670 @default.
• W3207570629 cites W2971273232 @default.
• W3207570629 cites W79394677 @default.
• W3207570629 cites W3037571487 @default.
• W3207570629 hasPublicationYear "2021" @default.
• W3207570629 type Work @default.
• W3207570629 sameAs 3207570629 @default.
• W3207570629 citedByCount "0" @default.
• W3207570629 crossrefType "posted-content" @default.
• W3207570629 hasAuthorship W3207570629A5000423552 @default.
• W3207570629 hasAuthorship W3207570629A5051794439 @default.
• W3207570629 hasAuthorship W3207570629A5065393757 @default.
• W3207570629 hasConcept C105795698 @default.
• W3207570629 hasConcept C111472728 @default.
• W3207570629 hasConcept C124304363 @default.
• W3207570629 hasConcept C13280743 @default.
• W3207570629 hasConcept C134306372 @default.
• W3207570629 hasConcept C138885662 @default.
• W3207570629 hasConcept C151730666 @default.
• W3207570629 hasConcept C154945302 @default.
• W3207570629 hasConcept C177264268 @default.
• W3207570629 hasConcept C184337299 @default.
• W3207570629 hasConcept C185798385 @default.
• W3207570629 hasConcept C199360897 @default.
• W3207570629 hasConcept C205649164 @default.
• W3207570629 hasConcept C2779343474 @default.
• W3207570629 hasConcept C33923547 @default.
• W3207570629 hasConcept C41008148 @default.
• W3207570629 hasConcept C72434380 @default.
• W3207570629 hasConcept C75306776 @default.
• W3207570629 hasConcept C80444323 @default.
• W3207570629 hasConcept C86803240 @default.
• W3207570629 hasConcept C97541855 @default.
• W3207570629 hasConceptScore W3207570629C105795698 @default.
• W3207570629 hasConceptScore W3207570629C111472728 @default.
• W3207570629 hasConceptScore W3207570629C124304363 @default.
• W3207570629 hasConceptScore W3207570629C13280743 @default.
• W3207570629 hasConceptScore W3207570629C134306372 @default.
• W3207570629 hasConceptScore W3207570629C138885662 @default.
• W3207570629 hasConceptScore W3207570629C151730666 @default.
• W3207570629 hasConceptScore W3207570629C154945302 @default.
• W3207570629 hasConceptScore W3207570629C177264268 @default.
• W3207570629 hasConceptScore W3207570629C184337299 @default.
• W3207570629 hasConceptScore W3207570629C185798385 @default.
• W3207570629 hasConceptScore W3207570629C199360897 @default.
• W3207570629 hasConceptScore W3207570629C205649164 @default.
• W3207570629 hasConceptScore W3207570629C2779343474 @default.
• W3207570629 hasConceptScore W3207570629C33923547 @default.
• W3207570629 hasConceptScore W3207570629C41008148 @default.
• W3207570629 hasConceptScore W3207570629C72434380 @default.
• W3207570629 hasConceptScore W3207570629C75306776 @default.
• W3207570629 hasConceptScore W3207570629C80444323 @default.
• W3207570629 hasConceptScore W3207570629C86803240 @default.
• W3207570629 hasConceptScore W3207570629C97541855 @default.
• W3207570629 hasLocation W32075706291 @default.
• W3207570629 hasOpenAccess W3207570629 @default.
• W3207570629 hasPrimaryLocation W32075706291 @default.
• W3207570629 hasRelatedWork W1503459987 @default.
• W3207570629 hasRelatedWork W1547065368 @default.
• W3207570629 hasRelatedWork W1560267378 @default.
• W3207570629 hasRelatedWork W158045570 @default.
• W3207570629 hasRelatedWork W1723461025 @default.
• W3207570629 hasRelatedWork W1761589367 @default.
• W3207570629 hasRelatedWork W1834134021 @default.
• W3207570629 hasRelatedWork W2021430251 @default.
• W3207570629 hasRelatedWork W2074731406 @default.
• W3207570629 hasRelatedWork W2083249903 @default.
• W3207570629 hasRelatedWork W2389007909 @default.
• W3207570629 hasRelatedWork W2397486648 @default.
• W3207570629 hasRelatedWork W2408475525 @default.
• W3207570629 hasRelatedWork W2561581089 @default.
• W3207570629 hasRelatedWork W2796623750 @default.

• next