Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313641521> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W4313641521 endingPage "317" @default.
- W4313641521 startingPage "306" @default.
- W4313641521 abstract "Reinforcement learning methods often produce brittle policies — policies that perform well during training, but generalize poorly beyond their direct training experience, thus becoming unstable under small disturbances. To address this issue, we propose a method for stabilizing a control policy in the space of configuration paths. It is applied post-training and relies purely on the data produced during training, as well as on an instantaneous control-matrix estimation. The approach is evaluated empirically on a planar bipedal walker subjected to a variety of perturbations. The control policies obtained via reinforcement learning are compared against their stabilized counterparts. Across different experiments, we find two- to four-fold increase in stability, when measured in terms of the perturbation amplitudes. We also provide a zero-dynamics interpretation of our approach." @default.
- W4313641521 created "2023-01-07" @default.
- W4313641521 creator A5070758620 @default.
- W4313641521 date "2023-01-01" @default.
- W4313641521 modified "2023-10-09" @default.
- W4313641521 title "Configuration Path Control" @default.
- W4313641521 cites W1504362584 @default.
- W4313641521 cites W1945123189 @default.
- W4313641521 cites W1975230295 @default.
- W4313641521 cites W1977655452 @default.
- W4313641521 cites W2103666082 @default.
- W4313641521 cites W2160815625 @default.
- W4313641521 cites W2261254692 @default.
- W4313641521 cites W2573784347 @default.
- W4313641521 cites W2575705757 @default.
- W4313641521 cites W2595845486 @default.
- W4313641521 cites W2605102758 @default.
- W4313641521 cites W2742229469 @default.
- W4313641521 cites W2773691349 @default.
- W4313641521 cites W2884001105 @default.
- W4313641521 cites W2905476369 @default.
- W4313641521 cites W2909553221 @default.
- W4313641521 cites W2919115771 @default.
- W4313641521 cites W2960705509 @default.
- W4313641521 cites W2962736495 @default.
- W4313641521 cites W2962957005 @default.
- W4313641521 cites W2972533062 @default.
- W4313641521 cites W2972798201 @default.
- W4313641521 cites W4243656494 @default.
- W4313641521 cites W4250058668 @default.
- W4313641521 cites W4300848695 @default.
- W4313641521 doi "https://doi.org/10.1007/s12555-021-0466-5" @default.
- W4313641521 hasPublicationYear "2023" @default.
- W4313641521 type Work @default.
- W4313641521 citedByCount "0" @default.
- W4313641521 crossrefType "journal-article" @default.
- W4313641521 hasAuthorship W4313641521A5070758620 @default.
- W4313641521 hasBestOaLocation W43136415212 @default.
- W4313641521 hasConcept C112972136 @default.
- W4313641521 hasConcept C119857082 @default.
- W4313641521 hasConcept C121332964 @default.
- W4313641521 hasConcept C136197465 @default.
- W4313641521 hasConcept C154945302 @default.
- W4313641521 hasConcept C177918212 @default.
- W4313641521 hasConcept C2775924081 @default.
- W4313641521 hasConcept C28704281 @default.
- W4313641521 hasConcept C33923547 @default.
- W4313641521 hasConcept C34413123 @default.
- W4313641521 hasConcept C41008148 @default.
- W4313641521 hasConcept C47446073 @default.
- W4313641521 hasConcept C62520636 @default.
- W4313641521 hasConcept C90509273 @default.
- W4313641521 hasConcept C97541855 @default.
- W4313641521 hasConceptScore W4313641521C112972136 @default.
- W4313641521 hasConceptScore W4313641521C119857082 @default.
- W4313641521 hasConceptScore W4313641521C121332964 @default.
- W4313641521 hasConceptScore W4313641521C136197465 @default.
- W4313641521 hasConceptScore W4313641521C154945302 @default.
- W4313641521 hasConceptScore W4313641521C177918212 @default.
- W4313641521 hasConceptScore W4313641521C2775924081 @default.
- W4313641521 hasConceptScore W4313641521C28704281 @default.
- W4313641521 hasConceptScore W4313641521C33923547 @default.
- W4313641521 hasConceptScore W4313641521C34413123 @default.
- W4313641521 hasConceptScore W4313641521C41008148 @default.
- W4313641521 hasConceptScore W4313641521C47446073 @default.
- W4313641521 hasConceptScore W4313641521C62520636 @default.
- W4313641521 hasConceptScore W4313641521C90509273 @default.
- W4313641521 hasConceptScore W4313641521C97541855 @default.
- W4313641521 hasIssue "1" @default.
- W4313641521 hasLocation W43136415211 @default.
- W4313641521 hasLocation W43136415212 @default.
- W4313641521 hasLocation W43136415213 @default.
- W4313641521 hasOpenAccess W4313641521 @default.
- W4313641521 hasPrimaryLocation W43136415211 @default.
- W4313641521 hasRelatedWork W2467617359 @default.
- W4313641521 hasRelatedWork W2475425029 @default.
- W4313641521 hasRelatedWork W2649651290 @default.
- W4313641521 hasRelatedWork W2670177905 @default.
- W4313641521 hasRelatedWork W2691078541 @default.
- W4313641521 hasRelatedWork W2691865310 @default.
- W4313641521 hasRelatedWork W2698589458 @default.
- W4313641521 hasRelatedWork W3074294383 @default.
- W4313641521 hasRelatedWork W4238798220 @default.
- W4313641521 hasRelatedWork W4319083788 @default.
- W4313641521 hasVolume "21" @default.
- W4313641521 isParatext "false" @default.
- W4313641521 isRetracted "false" @default.
- W4313641521 workType "article" @default.