FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3131034247> ?p ?o ?g. }

• W3131034247 abstract "While humans and animals learn incrementally during their lifetimes and exploit their experience to solve new tasks, standard deep reinforcement learning methods specialize to solve only one task at a time and, as a result, the information they acquire is hardly reusable in new situations. Here, we introduce a new perspective on the problem of leveraging prior knowledge to solve future unknown tasks. We show that learning discrete concept-like representations of sensory inputs can provide a high-level abstraction that is common across multiple tasks, thus facilitating the transference of information. In particular, we show that it is possible to learn such representations by self-supervision, following an information theoretic approach, and that they improve the sample efficiency by providing prior policies that guide the policy learning process. Our method is able to learn concepts in locomotive tasks that reduce the number of optimization steps in both known and unknown tasks, opening a new path to endow artificial agents with generalization abilities." @default.
• W3131034247 created "2021-03-01" @default.
• W3131034247 creator A5001455035 @default.
• W3131034247 creator A5010955102 @default.
• W3131034247 creator A5032148473 @default.
• W3131034247 date "2020-05-16" @default.
• W3131034247 modified "2023-09-27" @default.
• W3131034247 title "Learning Transferable Concepts in Deep Reinforcement Learning" @default.
• W3131034247 cites W1686946872 @default.
• W3131034247 cites W1738827650 @default.
• W3131034247 cites W1771410628 @default.
• W3131034247 cites W1994618660 @default.
• W3131034247 cites W2001685400 @default.
• W3131034247 cites W2014932765 @default.
• W3131034247 cites W2038794597 @default.
• W3131034247 cites W2052261430 @default.
• W3131034247 cites W2061982926 @default.
• W3131034247 cites W2065279029 @default.
• W3131034247 cites W2066481926 @default.
• W3131034247 cites W2073820635 @default.
• W3131034247 cites W2105055424 @default.
• W3131034247 cites W2109910161 @default.
• W3131034247 cites W2118917307 @default.
• W3131034247 cites W2119101994 @default.
• W3131034247 cites W2121863487 @default.
• W3131034247 cites W2122925692 @default.
• W3131034247 cites W2145339207 @default.
• W3131034247 cites W2158782408 @default.
• W3131034247 cites W2173564293 @default.
• W3131034247 cites W2273912701 @default.
• W3131034247 cites W2521862911 @default.
• W3131034247 cites W2726187156 @default.
• W3131034247 cites W2766447205 @default.
• W3131034247 cites W2798991696 @default.
• W3131034247 cites W2886442201 @default.
• W3131034247 cites W2887997457 @default.
• W3131034247 cites W2896084542 @default.
• W3131034247 cites W2904246096 @default.
• W3131034247 cites W2913403708 @default.
• W3131034247 cites W2916826721 @default.
• W3131034247 cites W2942538153 @default.
• W3131034247 cites W2948380112 @default.
• W3131034247 cites W2962861113 @default.
• W3131034247 cites W2963142324 @default.
• W3131034247 cites W2963161674 @default.
• W3131034247 cites W2963262099 @default.
• W3131034247 cites W2963438456 @default.
• W3131034247 cites W2963521490 @default.
• W3131034247 cites W2963641140 @default.
• W3131034247 cites W2963680188 @default.
• W3131034247 cites W2964009285 @default.
• W3131034247 cites W2964067469 @default.
• W3131034247 cites W2964118262 @default.
• W3131034247 cites W2964160479 @default.
• W3131034247 cites W2964174623 @default.
• W3131034247 cites W2964227312 @default.
• W3131034247 cites W2970214542 @default.
• W3131034247 cites W2971074500 @default.
• W3131034247 cites W2971202257 @default.
• W3131034247 cites W2976657239 @default.
• W3131034247 cites W2981344907 @default.
• W3131034247 cites W2982246413 @default.
• W3131034247 cites W2990747716 @default.
• W3131034247 cites W2995040055 @default.
• W3131034247 cites W2997250483 @default.
• W3131034247 cites W3005727454 @default.
• W3131034247 cites W3006604032 @default.
• W3131034247 cites W3031644095 @default.
• W3131034247 cites W3034419097 @default.
• W3131034247 cites W3037233728 @default.
• W3131034247 cites W3081674421 @default.
• W3131034247 cites W3092597801 @default.
• W3131034247 cites W3096964654 @default.
• W3131034247 cites W3106539628 @default.
• W3131034247 cites W3118210634 @default.
• W3131034247 cites W3122690883 @default.
• W3131034247 hasPublicationYear "2020" @default.
• W3131034247 type Work @default.
• W3131034247 sameAs 3131034247 @default.
• W3131034247 citedByCount "0" @default.
• W3131034247 crossrefType "posted-content" @default.
• W3131034247 hasAuthorship W3131034247A5001455035 @default.
• W3131034247 hasAuthorship W3131034247A5010955102 @default.
• W3131034247 hasAuthorship W3131034247A5032148473 @default.
• W3131034247 hasConcept C107457646 @default.
• W3131034247 hasConcept C111472728 @default.
• W3131034247 hasConcept C111919701 @default.
• W3131034247 hasConcept C119857082 @default.
• W3131034247 hasConcept C124304363 @default.
• W3131034247 hasConcept C12713177 @default.
• W3131034247 hasConcept C134306372 @default.
• W3131034247 hasConcept C138885662 @default.
• W3131034247 hasConcept C154945302 @default.
• W3131034247 hasConcept C162324750 @default.
• W3131034247 hasConcept C165696696 @default.
• W3131034247 hasConcept C177148314 @default.
• W3131034247 hasConcept C187736073 @default.
• W3131034247 hasConcept C199360897 @default.
• W3131034247 hasConcept C2777735758 @default.
• W3131034247 hasConcept C2778445095 @default.

• next