Matches in SemOpenAlex for { <https://semopenalex.org/work/W3132146241> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W3132146241 abstract "Deep Reinforcement Learning (DRL) can distill behavioural policies from sensory input that solve complex tasks, however, the policies tend to be task-specific and sample inefficient, requiring a large number of interactions with the environment that may be costly or impractical for many real world applications. Model-based DRL (MBRL) can allow learned behaviours and dynamics from one task to be translated to a new task in a related environment, but still suffer from low sample efficiency. In this work we introduce ReaPER, an algorithm that addresses the sample efficiency challenge in model-based DRL, we illustrate the power of the proposed solution on the DeepMind Control benchmark. Our improvements are driven by sparse , self-supervised, contrastive model representations and efficient use of past experience. We empirically analyze each novel component of ReaPER and analyze how they contribute to sample efficiency. We also illustrate how other standard alternatives fail to improve upon previous methods. Code will be made available." @default.
- W3132146241 created "2021-03-01" @default.
- W3132146241 creator A5008537744 @default.
- W3132146241 creator A5025218580 @default.
- W3132146241 creator A5051941774 @default.
- W3132146241 date "2021-05-04" @default.
- W3132146241 modified "2023-09-27" @default.
- W3132146241 title "ReaPER: Improving Sample Efficiency in Model-Based Latent Imagination" @default.
- W3132146241 hasPublicationYear "2021" @default.
- W3132146241 type Work @default.
- W3132146241 sameAs 3132146241 @default.
- W3132146241 citedByCount "0" @default.
- W3132146241 crossrefType "journal-article" @default.
- W3132146241 hasAuthorship W3132146241A5008537744 @default.
- W3132146241 hasAuthorship W3132146241A5025218580 @default.
- W3132146241 hasAuthorship W3132146241A5051941774 @default.
- W3132146241 hasConcept C119857082 @default.
- W3132146241 hasConcept C121332964 @default.
- W3132146241 hasConcept C127413603 @default.
- W3132146241 hasConcept C13280743 @default.
- W3132146241 hasConcept C154945302 @default.
- W3132146241 hasConcept C168167062 @default.
- W3132146241 hasConcept C177264268 @default.
- W3132146241 hasConcept C185592680 @default.
- W3132146241 hasConcept C185798385 @default.
- W3132146241 hasConcept C198531522 @default.
- W3132146241 hasConcept C199360897 @default.
- W3132146241 hasConcept C201995342 @default.
- W3132146241 hasConcept C205649164 @default.
- W3132146241 hasConcept C2776760102 @default.
- W3132146241 hasConcept C2780451532 @default.
- W3132146241 hasConcept C41008148 @default.
- W3132146241 hasConcept C43617362 @default.
- W3132146241 hasConcept C97355855 @default.
- W3132146241 hasConcept C97541855 @default.
- W3132146241 hasConceptScore W3132146241C119857082 @default.
- W3132146241 hasConceptScore W3132146241C121332964 @default.
- W3132146241 hasConceptScore W3132146241C127413603 @default.
- W3132146241 hasConceptScore W3132146241C13280743 @default.
- W3132146241 hasConceptScore W3132146241C154945302 @default.
- W3132146241 hasConceptScore W3132146241C168167062 @default.
- W3132146241 hasConceptScore W3132146241C177264268 @default.
- W3132146241 hasConceptScore W3132146241C185592680 @default.
- W3132146241 hasConceptScore W3132146241C185798385 @default.
- W3132146241 hasConceptScore W3132146241C198531522 @default.
- W3132146241 hasConceptScore W3132146241C199360897 @default.
- W3132146241 hasConceptScore W3132146241C201995342 @default.
- W3132146241 hasConceptScore W3132146241C205649164 @default.
- W3132146241 hasConceptScore W3132146241C2776760102 @default.
- W3132146241 hasConceptScore W3132146241C2780451532 @default.
- W3132146241 hasConceptScore W3132146241C41008148 @default.
- W3132146241 hasConceptScore W3132146241C43617362 @default.
- W3132146241 hasConceptScore W3132146241C97355855 @default.
- W3132146241 hasConceptScore W3132146241C97541855 @default.
- W3132146241 hasLocation W31321462411 @default.
- W3132146241 hasOpenAccess W3132146241 @default.
- W3132146241 hasPrimaryLocation W31321462411 @default.
- W3132146241 hasRelatedWork W2294805292 @default.
- W3132146241 hasRelatedWork W2626860042 @default.
- W3132146241 hasRelatedWork W2899041500 @default.
- W3132146241 hasRelatedWork W2907704766 @default.
- W3132146241 hasRelatedWork W2948198678 @default.
- W3132146241 hasRelatedWork W2952526277 @default.
- W3132146241 hasRelatedWork W2963199420 @default.
- W3132146241 hasRelatedWork W2968652061 @default.
- W3132146241 hasRelatedWork W3007369745 @default.
- W3132146241 hasRelatedWork W3014879845 @default.
- W3132146241 hasRelatedWork W3026304494 @default.
- W3132146241 hasRelatedWork W3034493393 @default.
- W3132146241 hasRelatedWork W3035216917 @default.
- W3132146241 hasRelatedWork W3043612295 @default.
- W3132146241 hasRelatedWork W3091395917 @default.
- W3132146241 hasRelatedWork W3103763075 @default.
- W3132146241 hasRelatedWork W3131310681 @default.
- W3132146241 hasRelatedWork W3175558129 @default.
- W3132146241 hasRelatedWork W3196302130 @default.
- W3132146241 hasRelatedWork W3213584136 @default.
- W3132146241 isParatext "false" @default.
- W3132146241 isRetracted "false" @default.
- W3132146241 magId "3132146241" @default.
- W3132146241 workType "article" @default.