Matches in SemOpenAlex for { <https://semopenalex.org/work/W2887933419> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W2887933419 endingPage "5377" @default.
- W2887933419 startingPage "5367" @default.
- W2887933419 abstract "Real-world applications often naturally decompose into several sub-tasks. In many settings (e.g., robotics) demonstrations provide a natural way to specify the sub-tasks. However, most methods for learning from demonstrations either do not provide guarantees that the artifacts learned for the sub-tasks can be safely recombined or limit the types of composition available. Motivated by this deficit, we consider the problem of inferring Boolean non-Markovian rewards (also known as logical trace properties or specifications) from demonstrations provided by an agent operating in an uncertain, stochastic environment. Crucially, specifications admit well-defined composition rules that are typically easy to interpret. In this paper, we formulate the specification inference task as a maximum a posteriori (MAP) probability inference problem, apply the principle of maximum entropy to derive an analytic demonstration likelihood model and give an efficient approach to search for the most likely specification in a large candidate pool of specifications. In our experiments, we demonstrate how learning specifications can help avoid common problems that often arise due to ad-hoc reward composition." @default.
- W2887933419 created "2018-08-22" @default.
- W2887933419 creator A5035902535 @default.
- W2887933419 creator A5044414209 @default.
- W2887933419 creator A5064230639 @default.
- W2887933419 creator A5083477814 @default.
- W2887933419 creator A5086308451 @default.
- W2887933419 date "2018-01-01" @default.
- W2887933419 modified "2023-09-22" @default.
- W2887933419 title "Learning Task Specifications from Demonstrations" @default.
- W2887933419 hasPublicationYear "2018" @default.
- W2887933419 type Work @default.
- W2887933419 sameAs 2887933419 @default.
- W2887933419 citedByCount "22" @default.
- W2887933419 countsByYear W28879334192016 @default.
- W2887933419 countsByYear W28879334192019 @default.
- W2887933419 countsByYear W28879334192020 @default.
- W2887933419 countsByYear W28879334192021 @default.
- W2887933419 crossrefType "proceedings-article" @default.
- W2887933419 hasAuthorship W2887933419A5035902535 @default.
- W2887933419 hasAuthorship W2887933419A5044414209 @default.
- W2887933419 hasAuthorship W2887933419A5064230639 @default.
- W2887933419 hasAuthorship W2887933419A5083477814 @default.
- W2887933419 hasAuthorship W2887933419A5086308451 @default.
- W2887933419 hasConcept C111472728 @default.
- W2887933419 hasConcept C119857082 @default.
- W2887933419 hasConcept C138885662 @default.
- W2887933419 hasConcept C154945302 @default.
- W2887933419 hasConcept C162324750 @default.
- W2887933419 hasConcept C187736073 @default.
- W2887933419 hasConcept C2776214188 @default.
- W2887933419 hasConcept C2780451532 @default.
- W2887933419 hasConcept C41008148 @default.
- W2887933419 hasConcept C41895202 @default.
- W2887933419 hasConcept C75291252 @default.
- W2887933419 hasConcept C75553542 @default.
- W2887933419 hasConcept C80444323 @default.
- W2887933419 hasConcept C9679016 @default.
- W2887933419 hasConceptScore W2887933419C111472728 @default.
- W2887933419 hasConceptScore W2887933419C119857082 @default.
- W2887933419 hasConceptScore W2887933419C138885662 @default.
- W2887933419 hasConceptScore W2887933419C154945302 @default.
- W2887933419 hasConceptScore W2887933419C162324750 @default.
- W2887933419 hasConceptScore W2887933419C187736073 @default.
- W2887933419 hasConceptScore W2887933419C2776214188 @default.
- W2887933419 hasConceptScore W2887933419C2780451532 @default.
- W2887933419 hasConceptScore W2887933419C41008148 @default.
- W2887933419 hasConceptScore W2887933419C41895202 @default.
- W2887933419 hasConceptScore W2887933419C75291252 @default.
- W2887933419 hasConceptScore W2887933419C75553542 @default.
- W2887933419 hasConceptScore W2887933419C80444323 @default.
- W2887933419 hasConceptScore W2887933419C9679016 @default.
- W2887933419 hasLocation W28879334191 @default.
- W2887933419 hasOpenAccess W2887933419 @default.
- W2887933419 hasPrimaryLocation W28879334191 @default.
- W2887933419 hasRelatedWork W1989445634 @default.
- W2887933419 hasRelatedWork W1999874108 @default.
- W2887933419 hasRelatedWork W2023808162 @default.
- W2887933419 hasRelatedWork W2032558547 @default.
- W2887933419 hasRelatedWork W2061562262 @default.
- W2887933419 hasRelatedWork W2098774185 @default.
- W2887933419 hasRelatedWork W2147544021 @default.
- W2887933419 hasRelatedWork W2169498096 @default.
- W2887933419 hasRelatedWork W2339807279 @default.
- W2887933419 hasRelatedWork W2474788619 @default.
- W2887933419 hasRelatedWork W2531762107 @default.
- W2887933419 hasRelatedWork W2567705466 @default.
- W2887933419 hasRelatedWork W2804948070 @default.
- W2887933419 hasRelatedWork W2807161818 @default.
- W2887933419 hasRelatedWork W2808386811 @default.
- W2887933419 hasRelatedWork W2889713213 @default.
- W2887933419 hasRelatedWork W2965569070 @default.
- W2887933419 hasRelatedWork W2966289416 @default.
- W2887933419 hasRelatedWork W2972698871 @default.
- W2887933419 hasRelatedWork W3100743289 @default.
- W2887933419 hasVolume "31" @default.
- W2887933419 isParatext "false" @default.
- W2887933419 isRetracted "false" @default.
- W2887933419 magId "2887933419" @default.
- W2887933419 workType "article" @default.