Matches in SemOpenAlex for { <https://semopenalex.org/work/W3196158767> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W3196158767 abstract "The success of deep reinforcement learning approaches to learn dexterous manipulation skills strongly hinges on the rewards assigned to actions during task execution. The usual approach is to handcraft the reward function but due to the high complexity of dexterous manipulations the reward definition demands large engineering effort for each particular task. To avoid this burden, we use an inverse reinforcement learning (IRL) approach to automatically learn the reward function using samples obtained from demonstrations of desired behaviours. We have identified that the learned rewards using existing IRL approaches are strongly biased towards demonstrated actions due to the scarcity of samples in the vast state-action space of dexterous manipulation applications. This significantly hinders performance due to unreliable reward estimations in regions unexplored during demonstration. We use statistical tools for random sample generation and reward normalization to reduce this bias. We show that this approach improves learning stability and transferability of IRL for dexterous manipulation tasks. Project page: https://sites.google.con view/irl-for-dexterous-hand" @default.
- W3196158767 created "2021-08-30" @default.
- W3196158767 creator A5051585666 @default.
- W3196158767 creator A5059129926 @default.
- W3196158767 creator A5088081800 @default.
- W3196158767 date "2021-08-23" @default.
- W3196158767 modified "2023-09-26" @default.
- W3196158767 title "Inverse reinforcement learning for dexterous hand manipulation" @default.
- W3196158767 cites W1564897360 @default.
- W3196158767 cites W2158782408 @default.
- W3196158767 cites W3103251932 @default.
- W3196158767 doi "https://doi.org/10.1109/icdl49984.2021.9515637" @default.
- W3196158767 hasPublicationYear "2021" @default.
- W3196158767 type Work @default.
- W3196158767 sameAs 3196158767 @default.
- W3196158767 citedByCount "4" @default.
- W3196158767 countsByYear W31961587672022 @default.
- W3196158767 countsByYear W31961587672023 @default.
- W3196158767 crossrefType "proceedings-article" @default.
- W3196158767 hasAuthorship W3196158767A5051585666 @default.
- W3196158767 hasAuthorship W3196158767A5059129926 @default.
- W3196158767 hasAuthorship W3196158767A5088081800 @default.
- W3196158767 hasConcept C107457646 @default.
- W3196158767 hasConcept C119857082 @default.
- W3196158767 hasConcept C127413603 @default.
- W3196158767 hasConcept C136886441 @default.
- W3196158767 hasConcept C14036430 @default.
- W3196158767 hasConcept C144024400 @default.
- W3196158767 hasConcept C154945302 @default.
- W3196158767 hasConcept C19165224 @default.
- W3196158767 hasConcept C201995342 @default.
- W3196158767 hasConcept C2780451532 @default.
- W3196158767 hasConcept C41008148 @default.
- W3196158767 hasConcept C66938386 @default.
- W3196158767 hasConcept C67203356 @default.
- W3196158767 hasConcept C78458016 @default.
- W3196158767 hasConcept C86803240 @default.
- W3196158767 hasConcept C97541855 @default.
- W3196158767 hasConceptScore W3196158767C107457646 @default.
- W3196158767 hasConceptScore W3196158767C119857082 @default.
- W3196158767 hasConceptScore W3196158767C127413603 @default.
- W3196158767 hasConceptScore W3196158767C136886441 @default.
- W3196158767 hasConceptScore W3196158767C14036430 @default.
- W3196158767 hasConceptScore W3196158767C144024400 @default.
- W3196158767 hasConceptScore W3196158767C154945302 @default.
- W3196158767 hasConceptScore W3196158767C19165224 @default.
- W3196158767 hasConceptScore W3196158767C201995342 @default.
- W3196158767 hasConceptScore W3196158767C2780451532 @default.
- W3196158767 hasConceptScore W3196158767C41008148 @default.
- W3196158767 hasConceptScore W3196158767C66938386 @default.
- W3196158767 hasConceptScore W3196158767C67203356 @default.
- W3196158767 hasConceptScore W3196158767C78458016 @default.
- W3196158767 hasConceptScore W3196158767C86803240 @default.
- W3196158767 hasConceptScore W3196158767C97541855 @default.
- W3196158767 hasFunder F4320321181 @default.
- W3196158767 hasLocation W31961587671 @default.
- W3196158767 hasOpenAccess W3196158767 @default.
- W3196158767 hasPrimaryLocation W31961587671 @default.
- W3196158767 hasRelatedWork W1562959674 @default.
- W3196158767 hasRelatedWork W2047937115 @default.
- W3196158767 hasRelatedWork W2734912394 @default.
- W3196158767 hasRelatedWork W2774891019 @default.
- W3196158767 hasRelatedWork W2784161548 @default.
- W3196158767 hasRelatedWork W2923653485 @default.
- W3196158767 hasRelatedWork W2957776456 @default.
- W3196158767 hasRelatedWork W3022038857 @default.
- W3196158767 hasRelatedWork W4287626175 @default.
- W3196158767 hasRelatedWork W4319083788 @default.
- W3196158767 isParatext "false" @default.
- W3196158767 isRetracted "false" @default.
- W3196158767 magId "3196158767" @default.
- W3196158767 workType "article" @default.