Matches in SemOpenAlex for { <https://semopenalex.org/work/W3035074656> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W3035074656 endingPage "9098" @default.
- W3035074656 startingPage "9088" @default.
- W3035074656 abstract "The high sample complexity of reinforcement learning challenges its use in practice. A promising approach is to quickly adapt pre-trained policies to new environments. Existing methods for this policy adaptation problem typically rely on domain randomization and meta-learning, by sampling from some distribution of target environments during pre-training, and thus face difficulty on out-of-distribution target environments. We propose new model-based mechanisms that are able to make online adaptation in unseen target environments, by combining ideas from no-regret online learning and adaptive control. We prove that the approach learns policies in the target environment that can quickly recover trajectories from the source environment, and establish the rate of convergence in general settings. We demonstrate the benefits of our approach for policy adaptation in a diverse set of continuous control tasks, achieving the performance of state-of-the-art methods with much lower sample complexity." @default.
- W3035074656 created "2020-06-19" @default.
- W3035074656 creator A5032266950 @default.
- W3035074656 creator A5040891381 @default.
- W3035074656 creator A5047083726 @default.
- W3035074656 creator A5081085178 @default.
- W3035074656 date "2020-07-12" @default.
- W3035074656 modified "2023-09-23" @default.
- W3035074656 title "Provably Efficient Model-based Policy Adaptation" @default.
- W3035074656 hasPublicationYear "2020" @default.
- W3035074656 type Work @default.
- W3035074656 sameAs 3035074656 @default.
- W3035074656 citedByCount "2" @default.
- W3035074656 countsByYear W30350746562021 @default.
- W3035074656 crossrefType "proceedings-article" @default.
- W3035074656 hasAuthorship W3035074656A5032266950 @default.
- W3035074656 hasAuthorship W3035074656A5040891381 @default.
- W3035074656 hasAuthorship W3035074656A5047083726 @default.
- W3035074656 hasAuthorship W3035074656A5081085178 @default.
- W3035074656 hasConcept C119857082 @default.
- W3035074656 hasConcept C120665830 @default.
- W3035074656 hasConcept C121332964 @default.
- W3035074656 hasConcept C134306372 @default.
- W3035074656 hasConcept C139807058 @default.
- W3035074656 hasConcept C154945302 @default.
- W3035074656 hasConcept C162324750 @default.
- W3035074656 hasConcept C177264268 @default.
- W3035074656 hasConcept C185592680 @default.
- W3035074656 hasConcept C198531522 @default.
- W3035074656 hasConcept C199360897 @default.
- W3035074656 hasConcept C2775924081 @default.
- W3035074656 hasConcept C2777303404 @default.
- W3035074656 hasConcept C33923547 @default.
- W3035074656 hasConcept C36503486 @default.
- W3035074656 hasConcept C41008148 @default.
- W3035074656 hasConcept C43617362 @default.
- W3035074656 hasConcept C50522688 @default.
- W3035074656 hasConcept C50817715 @default.
- W3035074656 hasConcept C97541855 @default.
- W3035074656 hasConceptScore W3035074656C119857082 @default.
- W3035074656 hasConceptScore W3035074656C120665830 @default.
- W3035074656 hasConceptScore W3035074656C121332964 @default.
- W3035074656 hasConceptScore W3035074656C134306372 @default.
- W3035074656 hasConceptScore W3035074656C139807058 @default.
- W3035074656 hasConceptScore W3035074656C154945302 @default.
- W3035074656 hasConceptScore W3035074656C162324750 @default.
- W3035074656 hasConceptScore W3035074656C177264268 @default.
- W3035074656 hasConceptScore W3035074656C185592680 @default.
- W3035074656 hasConceptScore W3035074656C198531522 @default.
- W3035074656 hasConceptScore W3035074656C199360897 @default.
- W3035074656 hasConceptScore W3035074656C2775924081 @default.
- W3035074656 hasConceptScore W3035074656C2777303404 @default.
- W3035074656 hasConceptScore W3035074656C33923547 @default.
- W3035074656 hasConceptScore W3035074656C36503486 @default.
- W3035074656 hasConceptScore W3035074656C41008148 @default.
- W3035074656 hasConceptScore W3035074656C43617362 @default.
- W3035074656 hasConceptScore W3035074656C50522688 @default.
- W3035074656 hasConceptScore W3035074656C50817715 @default.
- W3035074656 hasConceptScore W3035074656C97541855 @default.
- W3035074656 hasOpenAccess W3035074656 @default.
- W3035074656 hasRelatedWork W2626860042 @default.
- W3035074656 hasRelatedWork W2794757725 @default.
- W3035074656 hasRelatedWork W2903630557 @default.
- W3035074656 hasRelatedWork W2907704766 @default.
- W3035074656 hasRelatedWork W2963826726 @default.
- W3035074656 hasRelatedWork W2963859851 @default.
- W3035074656 hasRelatedWork W2968021416 @default.
- W3035074656 hasRelatedWork W2982795998 @default.
- W3035074656 hasRelatedWork W3003777383 @default.
- W3035074656 hasRelatedWork W3022124161 @default.
- W3035074656 hasRelatedWork W3034552332 @default.
- W3035074656 hasRelatedWork W3035216917 @default.
- W3035074656 hasRelatedWork W3035689006 @default.
- W3035074656 hasRelatedWork W3084024636 @default.
- W3035074656 hasRelatedWork W3087992196 @default.
- W3035074656 hasRelatedWork W3092185126 @default.
- W3035074656 hasRelatedWork W3131310681 @default.
- W3035074656 hasRelatedWork W3133498163 @default.
- W3035074656 hasRelatedWork W3163571701 @default.
- W3035074656 hasRelatedWork W778742492 @default.
- W3035074656 hasVolume "1" @default.
- W3035074656 isParatext "false" @default.
- W3035074656 isRetracted "false" @default.
- W3035074656 magId "3035074656" @default.
- W3035074656 workType "article" @default.