Matches in SemOpenAlex for { <https://semopenalex.org/work/W3097732461> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W3097732461 abstract "Temporal abstraction allows reinforcement learning agents to represent knowledge and develop strategies over different temporal scales. The option-critic framework has been demonstrated to learn temporally extended actions, represented as options, end-to-end in a model-free setting. However, feasibility of option-critic remains limited due to two major challenges, multiple options adopting very similar behavior, or a shrinking set of task relevant options. These occurrences not only void the need for temporal abstraction, they also affect performance. In this paper, we tackle these problems by learning a diverse set of options. We introduce an information-theoretic intrinsic reward, which augments the task reward, as well as a novel termination objective, in order to encourage behavioral diversity in the option set. We show empirically that our proposed method is capable of learning options end-to-end on several discrete and continuous control tasks, outperforms option-critic by a wide margin. Furthermore, we show that our approach sustainably generates robust, reusable, reliable and interpretable options, in contrast to option-critic." @default.
- W3097732461 created "2020-11-09" @default.
- W3097732461 creator A5065836447 @default.
- W3097732461 creator A5072603195 @default.
- W3097732461 date "2020-11-04" @default.
- W3097732461 modified "2023-09-27" @default.
- W3097732461 title "Diversity-Enriched Option-Critic." @default.
- W3097732461 cites W1556824961 @default.
- W3097732461 cites W1585861384 @default.
- W3097732461 cites W1777239053 @default.
- W3097732461 cites W1980516134 @default.
- W3097732461 cites W1993411524 @default.
- W3097732461 cites W199552065 @default.
- W3097732461 cites W2109910161 @default.
- W3097732461 cites W2119567691 @default.
- W3097732461 cites W2121517924 @default.
- W3097732461 cites W2121863487 @default.
- W3097732461 cites W2143435603 @default.
- W3097732461 cites W2158548602 @default.
- W3097732461 cites W2158782408 @default.
- W3097732461 cites W2164424353 @default.
- W3097732461 cites W2556477470 @default.
- W3097732461 cites W2594829461 @default.
- W3097732461 cites W2736601468 @default.
- W3097732461 cites W2746141389 @default.
- W3097732461 cites W2771734675 @default.
- W3097732461 cites W2781726626 @default.
- W3097732461 cites W2918091860 @default.
- W3097732461 cites W2963142324 @default.
- W3097732461 cites W2963160877 @default.
- W3097732461 cites W2963276097 @default.
- W3097732461 cites W2963438456 @default.
- W3097732461 cites W2964043796 @default.
- W3097732461 cites W2964096423 @default.
- W3097732461 cites W2964227312 @default.
- W3097732461 cites W2995636097 @default.
- W3097732461 cites W2997250483 @default.
- W3097732461 cites W2997289589 @default.
- W3097732461 hasPublicationYear "2020" @default.
- W3097732461 type Work @default.
- W3097732461 sameAs 3097732461 @default.
- W3097732461 citedByCount "1" @default.
- W3097732461 countsByYear W30977324612022 @default.
- W3097732461 crossrefType "posted-content" @default.
- W3097732461 hasAuthorship W3097732461A5065836447 @default.
- W3097732461 hasAuthorship W3097732461A5072603195 @default.
- W3097732461 hasConcept C111472728 @default.
- W3097732461 hasConcept C119857082 @default.
- W3097732461 hasConcept C124304363 @default.
- W3097732461 hasConcept C127413603 @default.
- W3097732461 hasConcept C138885662 @default.
- W3097732461 hasConcept C154945302 @default.
- W3097732461 hasConcept C177264268 @default.
- W3097732461 hasConcept C199360897 @default.
- W3097732461 hasConcept C201995342 @default.
- W3097732461 hasConcept C2780451532 @default.
- W3097732461 hasConcept C41008148 @default.
- W3097732461 hasConcept C774472 @default.
- W3097732461 hasConcept C97541855 @default.
- W3097732461 hasConceptScore W3097732461C111472728 @default.
- W3097732461 hasConceptScore W3097732461C119857082 @default.
- W3097732461 hasConceptScore W3097732461C124304363 @default.
- W3097732461 hasConceptScore W3097732461C127413603 @default.
- W3097732461 hasConceptScore W3097732461C138885662 @default.
- W3097732461 hasConceptScore W3097732461C154945302 @default.
- W3097732461 hasConceptScore W3097732461C177264268 @default.
- W3097732461 hasConceptScore W3097732461C199360897 @default.
- W3097732461 hasConceptScore W3097732461C201995342 @default.
- W3097732461 hasConceptScore W3097732461C2780451532 @default.
- W3097732461 hasConceptScore W3097732461C41008148 @default.
- W3097732461 hasConceptScore W3097732461C774472 @default.
- W3097732461 hasConceptScore W3097732461C97541855 @default.
- W3097732461 hasLocation W30977324611 @default.
- W3097732461 hasOpenAccess W3097732461 @default.
- W3097732461 hasPrimaryLocation W30977324611 @default.
- W3097732461 hasRelatedWork W1585861384 @default.
- W3097732461 hasRelatedWork W2102000945 @default.
- W3097732461 hasRelatedWork W2907704766 @default.
- W3097732461 hasRelatedWork W2950040888 @default.
- W3097732461 hasRelatedWork W2988122408 @default.
- W3097732461 hasRelatedWork W2995179464 @default.
- W3097732461 hasRelatedWork W2995363872 @default.
- W3097732461 hasRelatedWork W2996965557 @default.
- W3097732461 hasRelatedWork W3005607450 @default.
- W3097732461 hasRelatedWork W3037620198 @default.
- W3097732461 hasRelatedWork W3084024636 @default.
- W3097732461 hasRelatedWork W3092485320 @default.
- W3097732461 hasRelatedWork W3093503755 @default.
- W3097732461 hasRelatedWork W3129981047 @default.
- W3097732461 hasRelatedWork W3167983015 @default.
- W3097732461 hasRelatedWork W3168815054 @default.
- W3097732461 hasRelatedWork W3173031723 @default.
- W3097732461 hasRelatedWork W3200692531 @default.
- W3097732461 hasRelatedWork W3200996868 @default.
- W3097732461 hasRelatedWork W3208244472 @default.
- W3097732461 isParatext "false" @default.
- W3097732461 isRetracted "false" @default.
- W3097732461 magId "3097732461" @default.
- W3097732461 workType "article" @default.