Matches in SemOpenAlex for { <https://semopenalex.org/work/W4302012168> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4302012168 abstract "End-to-end reinforcement learning techniques are among the most successful methods for robotic manipulation tasks. However, the training time required to find a good policy capable of solving complex tasks is prohibitively large. Therefore, depending on the computing resources available, it might not be feasible to use such techniques. The use of domain knowledge to decompose manipulation tasks into primitive skills, to be performed in sequence, could reduce the overall complexity of the learning problem, and hence reduce the amount of training required to achieve dexterity. In this paper, we propose the use of Davenport chained rotations to decompose complex 3D rotation goals into a concatenation of a smaller set of more simple rotation skills. State-of-the-art reinforcement-learning-based methods can then be trained using less overall simulated experience. We compare its performance with the popular Hindsight Experience Replay method, trained in an end-to-end fashion using the same amount of experience in a simulated robotic hand environment. Despite a general decrease in performance of the primitive skills when being sequentially executed, we find that decomposing arbitrary 3D rotations into elementary rotations is beneficial when computing resources are limited, obtaining increases of success rates of approximately 10% on the most complex 3D rotations with respect to the success rates obtained by HER trained in an end-to-end fashion, and increases of success rates between 20% and 40% on the most simple rotations." @default.
- W4302012168 created "2022-10-06" @default.
- W4302012168 creator A5005226665 @default.
- W4302012168 creator A5013317847 @default.
- W4302012168 creator A5030863883 @default.
- W4302012168 creator A5073924795 @default.
- W4302012168 creator A5074319083 @default.
- W4302012168 creator A5081902636 @default.
- W4302012168 date "2022-10-03" @default.
- W4302012168 modified "2023-10-05" @default.
- W4302012168 title "Hierarchical reinforcement learning for in-hand robotic manipulation using Davenport chained rotations" @default.
- W4302012168 doi "https://doi.org/10.48550/arxiv.2210.00795" @default.
- W4302012168 hasPublicationYear "2022" @default.
- W4302012168 type Work @default.
- W4302012168 citedByCount "0" @default.
- W4302012168 crossrefType "posted-content" @default.
- W4302012168 hasAuthorship W4302012168A5005226665 @default.
- W4302012168 hasAuthorship W4302012168A5013317847 @default.
- W4302012168 hasAuthorship W4302012168A5030863883 @default.
- W4302012168 hasAuthorship W4302012168A5073924795 @default.
- W4302012168 hasAuthorship W4302012168A5074319083 @default.
- W4302012168 hasAuthorship W4302012168A5081902636 @default.
- W4302012168 hasBestOaLocation W43020121681 @default.
- W4302012168 hasConcept C10347200 @default.
- W4302012168 hasConcept C107457646 @default.
- W4302012168 hasConcept C111472728 @default.
- W4302012168 hasConcept C119857082 @default.
- W4302012168 hasConcept C134306372 @default.
- W4302012168 hasConcept C138885662 @default.
- W4302012168 hasConcept C154945302 @default.
- W4302012168 hasConcept C15744967 @default.
- W4302012168 hasConcept C177264268 @default.
- W4302012168 hasConcept C180747234 @default.
- W4302012168 hasConcept C199360897 @default.
- W4302012168 hasConcept C2778112365 @default.
- W4302012168 hasConcept C2780586882 @default.
- W4302012168 hasConcept C33923547 @default.
- W4302012168 hasConcept C36503486 @default.
- W4302012168 hasConcept C41008148 @default.
- W4302012168 hasConcept C54355233 @default.
- W4302012168 hasConcept C74050887 @default.
- W4302012168 hasConcept C86803240 @default.
- W4302012168 hasConcept C87619178 @default.
- W4302012168 hasConcept C90509273 @default.
- W4302012168 hasConcept C94375191 @default.
- W4302012168 hasConcept C97541855 @default.
- W4302012168 hasConceptScore W4302012168C10347200 @default.
- W4302012168 hasConceptScore W4302012168C107457646 @default.
- W4302012168 hasConceptScore W4302012168C111472728 @default.
- W4302012168 hasConceptScore W4302012168C119857082 @default.
- W4302012168 hasConceptScore W4302012168C134306372 @default.
- W4302012168 hasConceptScore W4302012168C138885662 @default.
- W4302012168 hasConceptScore W4302012168C154945302 @default.
- W4302012168 hasConceptScore W4302012168C15744967 @default.
- W4302012168 hasConceptScore W4302012168C177264268 @default.
- W4302012168 hasConceptScore W4302012168C180747234 @default.
- W4302012168 hasConceptScore W4302012168C199360897 @default.
- W4302012168 hasConceptScore W4302012168C2778112365 @default.
- W4302012168 hasConceptScore W4302012168C2780586882 @default.
- W4302012168 hasConceptScore W4302012168C33923547 @default.
- W4302012168 hasConceptScore W4302012168C36503486 @default.
- W4302012168 hasConceptScore W4302012168C41008148 @default.
- W4302012168 hasConceptScore W4302012168C54355233 @default.
- W4302012168 hasConceptScore W4302012168C74050887 @default.
- W4302012168 hasConceptScore W4302012168C86803240 @default.
- W4302012168 hasConceptScore W4302012168C87619178 @default.
- W4302012168 hasConceptScore W4302012168C90509273 @default.
- W4302012168 hasConceptScore W4302012168C94375191 @default.
- W4302012168 hasConceptScore W4302012168C97541855 @default.
- W4302012168 hasLocation W43020121681 @default.
- W4302012168 hasLocation W43020121682 @default.
- W4302012168 hasOpenAccess W4302012168 @default.
- W4302012168 hasPrimaryLocation W43020121681 @default.
- W4302012168 hasRelatedWork W2011276890 @default.
- W4302012168 hasRelatedWork W2804672169 @default.
- W4302012168 hasRelatedWork W2890406131 @default.
- W4302012168 hasRelatedWork W3012552522 @default.
- W4302012168 hasRelatedWork W3173051288 @default.
- W4302012168 hasRelatedWork W3197854638 @default.
- W4302012168 hasRelatedWork W4225749814 @default.
- W4302012168 hasRelatedWork W4226336685 @default.
- W4302012168 hasRelatedWork W4295352814 @default.
- W4302012168 hasRelatedWork W4319083788 @default.
- W4302012168 isParatext "false" @default.
- W4302012168 isRetracted "false" @default.
- W4302012168 workType "article" @default.