FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206039721> ?p ?o ?g. }

• W3206039721 abstract "Reading to act is a prevalent but challenging task which requires the ability to reason from a concise instruction. However, previous works face the semantic mismatch between the low-level actions and the high-level language descriptions and require the human-designed curriculum to work properly. In this paper, we present a Feudal Reinforcement Learning (FRL) model consisting of a manager agent and a worker agent. The manager agent is a multi-hop plan generator dealing with high-level abstract information and generating a series of sub-goals in a backward manner. The worker agent deals with the low-level perceptions and actions to achieve the sub-goals one by one. In comparison, our FRL model effectively alleviate the mismatching between text-level inference and low-level perceptions and actions; and is general to various forms of environments, instructions and manuals; and our multi-hop plan generator can significantly boost for challenging tasks where multi-step reasoning form the texts is critical to resolve the instructed goals. We showcase our approach achieves competitive performance on two challenging tasks, Read to Fight Monsters (RTFM) and Messenger, without human-designed curriculum learning." @default.
• W3206039721 created "2021-10-25" @default.
• W3206039721 creator A5002072267 @default.
• W3206039721 creator A5021802340 @default.
• W3206039721 creator A5027142928 @default.
• W3206039721 creator A5039086633 @default.
• W3206039721 date "2021-10-13" @default.
• W3206039721 modified "2023-09-26" @default.
• W3206039721 title "Feudal Reinforcement Learning by Reading Manuals." @default.
• W3206039721 cites W1544827683 @default.
• W3206039721 cites W1610678877 @default.
• W3206039721 cites W1646707810 @default.
• W3206039721 cites W2064675550 @default.
• W3206039721 cites W2122223050 @default.
• W3206039721 cites W2125436846 @default.
• W3206039721 cites W2131774270 @default.
• W3206039721 cites W2160371091 @default.
• W3206039721 cites W2310425190 @default.
• W3206039721 cites W2594829461 @default.
• W3206039721 cites W2627585944 @default.
• W3206039721 cites W2739691807 @default.
• W3206039721 cites W2788448041 @default.
• W3206039721 cites W2799187742 @default.
• W3206039721 cites W2803281228 @default.
• W3206039721 cites W2810346659 @default.
• W3206039721 cites W2949241965 @default.
• W3206039721 cites W2949876402 @default.
• W3206039721 cites W2954579883 @default.
• W3206039721 cites W2963068985 @default.
• W3206039721 cites W2963306198 @default.
• W3206039721 cites W2963748441 @default.
• W3206039721 cites W2963800628 @default.
• W3206039721 cites W2963855730 @default.
• W3206039721 cites W2963866616 @default.
• W3206039721 cites W2963963993 @default.
• W3206039721 cites W2963985863 @default.
• W3206039721 cites W2964121744 @default.
• W3206039721 cites W2964935470 @default.
• W3206039721 cites W2970780738 @default.
• W3206039721 cites W2980077985 @default.
• W3206039721 cites W2982311130 @default.
• W3206039721 cites W2995264919 @default.
• W3206039721 cites W2996848635 @default.
• W3206039721 cites W2998557583 @default.
• W3206039721 cites W3003205975 @default.
• W3206039721 cites W3099954076 @default.
• W3206039721 cites W3122267274 @default.
• W3206039721 cites W3139434402 @default.
• W3206039721 cites W3202505567 @default.
• W3206039721 hasPublicationYear "2021" @default.
• W3206039721 type Work @default.
• W3206039721 sameAs 3206039721 @default.
• W3206039721 citedByCount "0" @default.
• W3206039721 crossrefType "posted-content" @default.
• W3206039721 hasAuthorship W3206039721A5002072267 @default.
• W3206039721 hasAuthorship W3206039721A5021802340 @default.
• W3206039721 hasAuthorship W3206039721A5027142928 @default.
• W3206039721 hasAuthorship W3206039721A5039086633 @default.
• W3206039721 hasConcept C138885662 @default.
• W3206039721 hasConcept C154945302 @default.
• W3206039721 hasConcept C15744967 @default.
• W3206039721 hasConcept C162324750 @default.
• W3206039721 hasConcept C166957645 @default.
• W3206039721 hasConcept C169760540 @default.
• W3206039721 hasConcept C17744445 @default.
• W3206039721 hasConcept C187736073 @default.
• W3206039721 hasConcept C19417346 @default.
• W3206039721 hasConcept C199539241 @default.
• W3206039721 hasConcept C26760741 @default.
• W3206039721 hasConcept C2776214188 @default.
• W3206039721 hasConcept C2776505523 @default.
• W3206039721 hasConcept C2780451532 @default.
• W3206039721 hasConcept C41008148 @default.
• W3206039721 hasConcept C41895202 @default.
• W3206039721 hasConcept C47177190 @default.
• W3206039721 hasConcept C511693568 @default.
• W3206039721 hasConcept C554936623 @default.
• W3206039721 hasConcept C94625758 @default.
• W3206039721 hasConcept C95457728 @default.
• W3206039721 hasConcept C97541855 @default.
• W3206039721 hasConceptScore W3206039721C138885662 @default.
• W3206039721 hasConceptScore W3206039721C154945302 @default.
• W3206039721 hasConceptScore W3206039721C15744967 @default.
• W3206039721 hasConceptScore W3206039721C162324750 @default.
• W3206039721 hasConceptScore W3206039721C166957645 @default.
• W3206039721 hasConceptScore W3206039721C169760540 @default.
• W3206039721 hasConceptScore W3206039721C17744445 @default.
• W3206039721 hasConceptScore W3206039721C187736073 @default.
• W3206039721 hasConceptScore W3206039721C19417346 @default.
• W3206039721 hasConceptScore W3206039721C199539241 @default.
• W3206039721 hasConceptScore W3206039721C26760741 @default.
• W3206039721 hasConceptScore W3206039721C2776214188 @default.
• W3206039721 hasConceptScore W3206039721C2776505523 @default.
• W3206039721 hasConceptScore W3206039721C2780451532 @default.
• W3206039721 hasConceptScore W3206039721C41008148 @default.
• W3206039721 hasConceptScore W3206039721C41895202 @default.
• W3206039721 hasConceptScore W3206039721C47177190 @default.
• W3206039721 hasConceptScore W3206039721C511693568 @default.
• W3206039721 hasConceptScore W3206039721C554936623 @default.
• W3206039721 hasConceptScore W3206039721C94625758 @default.

• next