Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951400270> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2951400270 abstract "We propose a method for tackling catastrophic forgetting in deep reinforcement learning that is textit{agnostic} to the timescale of changes in the distribution of experiences, does not require knowledge of task boundaries, and can adapt in textit{continuously} changing environments. In our textit{policy consolidation} model, the policy network interacts with a cascade of hidden networks that simultaneously remember the agent's policy at a range of timescales and regularise the current policy by its own history, thereby improving its ability to learn without forgetting. We find that the model improves continual learning relative to baselines on a number of continuous control tasks in single-task, alternating two-task, and multi-agent competitive self-play settings." @default.
- W2951400270 created "2019-06-27" @default.
- W2951400270 creator A5039497694 @default.
- W2951400270 creator A5065295907 @default.
- W2951400270 creator A5072322524 @default.
- W2951400270 date "2019-02-01" @default.
- W2951400270 modified "2023-09-27" @default.
- W2951400270 title "Policy Consolidation for Continual Reinforcement Learning" @default.
- W2951400270 cites W1191599655 @default.
- W2951400270 cites W1584431645 @default.
- W2951400270 cites W1682403713 @default.
- W2951400270 cites W1821462560 @default.
- W2951400270 cites W1889619166 @default.
- W2951400270 cites W2083133776 @default.
- W2951400270 cites W2165310684 @default.
- W2951400270 cites W2173248099 @default.
- W2951400270 cites W2257979135 @default.
- W2951400270 cites W2291986326 @default.
- W2951400270 cites W2415865124 @default.
- W2951400270 cites W2473930607 @default.
- W2951400270 cites W2529605558 @default.
- W2951400270 cites W2553665199 @default.
- W2951400270 cites W2560647685 @default.
- W2951400270 cites W2736601468 @default.
- W2951400270 cites W2737492962 @default.
- W2951400270 cites W2762872434 @default.
- W2951400270 cites W2786465559 @default.
- W2951400270 cites W2962724315 @default.
- W2951400270 cites W2963199420 @default.
- W2951400270 cites W2963559848 @default.
- W2951400270 cites W2963637944 @default.
- W2951400270 cites W2963850662 @default.
- W2951400270 cites W2964048876 @default.
- W2951400270 cites W2964088867 @default.
- W2951400270 cites W2970586779 @default.
- W2951400270 hasPublicationYear "2019" @default.
- W2951400270 type Work @default.
- W2951400270 sameAs 2951400270 @default.
- W2951400270 citedByCount "7" @default.
- W2951400270 countsByYear W29514002702019 @default.
- W2951400270 countsByYear W29514002702020 @default.
- W2951400270 countsByYear W29514002702021 @default.
- W2951400270 countsByYear W29514002702022 @default.
- W2951400270 crossrefType "posted-content" @default.
- W2951400270 hasAuthorship W2951400270A5039497694 @default.
- W2951400270 hasAuthorship W2951400270A5065295907 @default.
- W2951400270 hasAuthorship W2951400270A5072322524 @default.
- W2951400270 hasConcept C121955636 @default.
- W2951400270 hasConcept C154945302 @default.
- W2951400270 hasConcept C15744967 @default.
- W2951400270 hasConcept C162324750 @default.
- W2951400270 hasConcept C180747234 @default.
- W2951400270 hasConcept C187736073 @default.
- W2951400270 hasConcept C2776014549 @default.
- W2951400270 hasConcept C2780451532 @default.
- W2951400270 hasConcept C41008148 @default.
- W2951400270 hasConcept C7149132 @default.
- W2951400270 hasConcept C97541855 @default.
- W2951400270 hasConceptScore W2951400270C121955636 @default.
- W2951400270 hasConceptScore W2951400270C154945302 @default.
- W2951400270 hasConceptScore W2951400270C15744967 @default.
- W2951400270 hasConceptScore W2951400270C162324750 @default.
- W2951400270 hasConceptScore W2951400270C180747234 @default.
- W2951400270 hasConceptScore W2951400270C187736073 @default.
- W2951400270 hasConceptScore W2951400270C2776014549 @default.
- W2951400270 hasConceptScore W2951400270C2780451532 @default.
- W2951400270 hasConceptScore W2951400270C41008148 @default.
- W2951400270 hasConceptScore W2951400270C7149132 @default.
- W2951400270 hasConceptScore W2951400270C97541855 @default.
- W2951400270 hasLocation W29514002701 @default.
- W2951400270 hasOpenAccess W2951400270 @default.
- W2951400270 hasPrimaryLocation W29514002701 @default.
- W2951400270 hasRelatedWork W1534480106 @default.
- W2951400270 hasRelatedWork W2560647685 @default.
- W2951400270 hasRelatedWork W2605102581 @default.
- W2951400270 hasRelatedWork W2788388592 @default.
- W2951400270 hasRelatedWork W2897007337 @default.
- W2951400270 hasRelatedWork W2909911836 @default.
- W2951400270 hasRelatedWork W2918932727 @default.
- W2951400270 hasRelatedWork W2962724315 @default.
- W2951400270 hasRelatedWork W2964048876 @default.
- W2951400270 hasRelatedWork W2970586779 @default.
- W2951400270 hasRelatedWork W2981499914 @default.
- W2951400270 hasRelatedWork W2998135952 @default.
- W2951400270 hasRelatedWork W3103724573 @default.
- W2951400270 hasRelatedWork W3103936143 @default.
- W2951400270 hasRelatedWork W3111013878 @default.
- W2951400270 hasRelatedWork W3111937515 @default.
- W2951400270 hasRelatedWork W3124858998 @default.
- W2951400270 hasRelatedWork W3200711725 @default.
- W2951400270 hasRelatedWork W3213234780 @default.
- W2951400270 hasRelatedWork W3096684011 @default.
- W2951400270 isParatext "false" @default.
- W2951400270 isRetracted "false" @default.
- W2951400270 magId "2951400270" @default.
- W2951400270 workType "article" @default.