Matches in SemOpenAlex for { <https://semopenalex.org/work/W3168354865> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3168354865 endingPage "302" @default.
- W3168354865 startingPage "291" @default.
- W3168354865 abstract "Policy search reinforcement learning has been drawing much attention as a method of learning a robot control policy. In particular, policy search using such non-parametric policies as Gaussian process regression can learn optimal actions with high-dimensional and redundant sensors as input. However, previous methods implicitly assume that the optimal action becomes unique for each state. This assumption can severely limit such practical applications as robot manipulations since designing a reward function that appears in only one optimal action for complex tasks is difficult. The previous methods might have caused critical performance deterioration because the typical non-parametric policies cannot capture the optimal actions due to their unimodality. We propose novel approaches in non-parametric policy searches with multiple optimal actions and offer two different algorithms commonly based on a sparse Gaussian process prior and variational Bayesian inference. The following are the key ideas: (1) multimodality for capturing multiple optimal actions and (2) mode-seeking for capturing one optimal action by ignoring the others. First, we propose a multimodal sparse Gaussian process policy search that uses multiple overlapped GPs as a prior. Second, we propose a mode-seeking sparse Gaussian process policy search that uses the student-t distribution for a likelihood function. The effectiveness of those algorithms is demonstrated through applications to object manipulation tasks with multiple optimal actions in simulations." @default.
- W3168354865 created "2021-06-22" @default.
- W3168354865 creator A5018794368 @default.
- W3168354865 creator A5042074952 @default.
- W3168354865 date "2021-11-01" @default.
- W3168354865 modified "2023-10-18" @default.
- W3168354865 title "Variational policy search using sparse Gaussian process priors for learning multimodal optimal actions" @default.
- W3168354865 cites W1972063518 @default.
- W3168354865 cites W1973102501 @default.
- W3168354865 cites W1973310094 @default.
- W3168354865 cites W1977655452 @default.
- W3168354865 cites W2114355534 @default.
- W3168354865 cites W2145339207 @default.
- W3168354865 cites W2900582619 @default.
- W3168354865 doi "https://doi.org/10.1016/j.neunet.2021.06.010" @default.
- W3168354865 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34166892" @default.
- W3168354865 hasPublicationYear "2021" @default.
- W3168354865 type Work @default.
- W3168354865 sameAs 3168354865 @default.
- W3168354865 citedByCount "4" @default.
- W3168354865 countsByYear W31683548652022 @default.
- W3168354865 countsByYear W31683548652023 @default.
- W3168354865 crossrefType "journal-article" @default.
- W3168354865 hasAuthorship W3168354865A5018794368 @default.
- W3168354865 hasAuthorship W3168354865A5042074952 @default.
- W3168354865 hasBestOaLocation W31683548652 @default.
- W3168354865 hasConcept C105795698 @default.
- W3168354865 hasConcept C107673813 @default.
- W3168354865 hasConcept C111919701 @default.
- W3168354865 hasConcept C117251300 @default.
- W3168354865 hasConcept C119857082 @default.
- W3168354865 hasConcept C121332964 @default.
- W3168354865 hasConcept C126255220 @default.
- W3168354865 hasConcept C14036430 @default.
- W3168354865 hasConcept C154945302 @default.
- W3168354865 hasConcept C163716315 @default.
- W3168354865 hasConcept C177769412 @default.
- W3168354865 hasConcept C2780791683 @default.
- W3168354865 hasConcept C33923547 @default.
- W3168354865 hasConcept C41008148 @default.
- W3168354865 hasConcept C61326573 @default.
- W3168354865 hasConcept C62520636 @default.
- W3168354865 hasConcept C78458016 @default.
- W3168354865 hasConcept C86803240 @default.
- W3168354865 hasConcept C97541855 @default.
- W3168354865 hasConcept C98045186 @default.
- W3168354865 hasConceptScore W3168354865C105795698 @default.
- W3168354865 hasConceptScore W3168354865C107673813 @default.
- W3168354865 hasConceptScore W3168354865C111919701 @default.
- W3168354865 hasConceptScore W3168354865C117251300 @default.
- W3168354865 hasConceptScore W3168354865C119857082 @default.
- W3168354865 hasConceptScore W3168354865C121332964 @default.
- W3168354865 hasConceptScore W3168354865C126255220 @default.
- W3168354865 hasConceptScore W3168354865C14036430 @default.
- W3168354865 hasConceptScore W3168354865C154945302 @default.
- W3168354865 hasConceptScore W3168354865C163716315 @default.
- W3168354865 hasConceptScore W3168354865C177769412 @default.
- W3168354865 hasConceptScore W3168354865C2780791683 @default.
- W3168354865 hasConceptScore W3168354865C33923547 @default.
- W3168354865 hasConceptScore W3168354865C41008148 @default.
- W3168354865 hasConceptScore W3168354865C61326573 @default.
- W3168354865 hasConceptScore W3168354865C62520636 @default.
- W3168354865 hasConceptScore W3168354865C78458016 @default.
- W3168354865 hasConceptScore W3168354865C86803240 @default.
- W3168354865 hasConceptScore W3168354865C97541855 @default.
- W3168354865 hasConceptScore W3168354865C98045186 @default.
- W3168354865 hasLocation W31683548651 @default.
- W3168354865 hasLocation W31683548652 @default.
- W3168354865 hasOpenAccess W3168354865 @default.
- W3168354865 hasPrimaryLocation W31683548651 @default.
- W3168354865 hasRelatedWork W115891841 @default.
- W3168354865 hasRelatedWork W2074236736 @default.
- W3168354865 hasRelatedWork W2079291356 @default.
- W3168354865 hasRelatedWork W2122071569 @default.
- W3168354865 hasRelatedWork W2143042284 @default.
- W3168354865 hasRelatedWork W2144536756 @default.
- W3168354865 hasRelatedWork W2251221343 @default.
- W3168354865 hasRelatedWork W3022038857 @default.
- W3168354865 hasRelatedWork W4319083788 @default.
- W3168354865 hasRelatedWork W4319293735 @default.
- W3168354865 hasVolume "143" @default.
- W3168354865 isParatext "false" @default.
- W3168354865 isRetracted "false" @default.
- W3168354865 magId "3168354865" @default.
- W3168354865 workType "article" @default.