Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288414141> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W4288414141 abstract "As reinforcement learning agents are tasked with solving more challenging and diverse tasks, the ability to incorporate prior knowledge into the learning system and to exploit reusable structure in solution space is likely to become increasingly important. The KL-regularized expected reward objective constitutes one possible tool to this end. It introduces an additional component, a default or prior behavior, which can be learned alongside the policy and as such partially transforms the reinforcement learning problem into one of behavior modelling. In this work we consider the implications of this framework in cases where both the policy and default behavior are augmented with latent variables. We discuss how the resulting hierarchical structures can be used to implement different inductive biases and how their modularity can benefit transfer. Empirically we find that they can lead to faster learning and transfer on a range of continuous control tasks." @default.
- W4288414141 created "2022-07-29" @default.
- W4288414141 creator A5005912318 @default.
- W4288414141 creator A5013028446 @default.
- W4288414141 creator A5014567358 @default.
- W4288414141 creator A5039426831 @default.
- W4288414141 creator A5043910056 @default.
- W4288414141 creator A5062951341 @default.
- W4288414141 creator A5064373793 @default.
- W4288414141 creator A5067400056 @default.
- W4288414141 creator A5084673940 @default.
- W4288414141 date "2019-03-18" @default.
- W4288414141 modified "2023-09-23" @default.
- W4288414141 title "Exploiting Hierarchy for Learning and Transfer in KL-regularized RL" @default.
- W4288414141 doi "https://doi.org/10.48550/arxiv.1903.07438" @default.
- W4288414141 hasPublicationYear "2019" @default.
- W4288414141 type Work @default.
- W4288414141 citedByCount "0" @default.
- W4288414141 crossrefType "posted-content" @default.
- W4288414141 hasAuthorship W4288414141A5005912318 @default.
- W4288414141 hasAuthorship W4288414141A5013028446 @default.
- W4288414141 hasAuthorship W4288414141A5014567358 @default.
- W4288414141 hasAuthorship W4288414141A5039426831 @default.
- W4288414141 hasAuthorship W4288414141A5043910056 @default.
- W4288414141 hasAuthorship W4288414141A5062951341 @default.
- W4288414141 hasAuthorship W4288414141A5064373793 @default.
- W4288414141 hasAuthorship W4288414141A5067400056 @default.
- W4288414141 hasAuthorship W4288414141A5084673940 @default.
- W4288414141 hasBestOaLocation W42884141411 @default.
- W4288414141 hasConcept C111919701 @default.
- W4288414141 hasConcept C119857082 @default.
- W4288414141 hasConcept C121332964 @default.
- W4288414141 hasConcept C150899416 @default.
- W4288414141 hasConcept C154945302 @default.
- W4288414141 hasConcept C162324750 @default.
- W4288414141 hasConcept C165696696 @default.
- W4288414141 hasConcept C168167062 @default.
- W4288414141 hasConcept C2778572836 @default.
- W4288414141 hasConcept C2779478453 @default.
- W4288414141 hasConcept C31170391 @default.
- W4288414141 hasConcept C34447519 @default.
- W4288414141 hasConcept C38652104 @default.
- W4288414141 hasConcept C41008148 @default.
- W4288414141 hasConcept C54355233 @default.
- W4288414141 hasConcept C86803240 @default.
- W4288414141 hasConcept C97355855 @default.
- W4288414141 hasConcept C97541855 @default.
- W4288414141 hasConceptScore W4288414141C111919701 @default.
- W4288414141 hasConceptScore W4288414141C119857082 @default.
- W4288414141 hasConceptScore W4288414141C121332964 @default.
- W4288414141 hasConceptScore W4288414141C150899416 @default.
- W4288414141 hasConceptScore W4288414141C154945302 @default.
- W4288414141 hasConceptScore W4288414141C162324750 @default.
- W4288414141 hasConceptScore W4288414141C165696696 @default.
- W4288414141 hasConceptScore W4288414141C168167062 @default.
- W4288414141 hasConceptScore W4288414141C2778572836 @default.
- W4288414141 hasConceptScore W4288414141C2779478453 @default.
- W4288414141 hasConceptScore W4288414141C31170391 @default.
- W4288414141 hasConceptScore W4288414141C34447519 @default.
- W4288414141 hasConceptScore W4288414141C38652104 @default.
- W4288414141 hasConceptScore W4288414141C41008148 @default.
- W4288414141 hasConceptScore W4288414141C54355233 @default.
- W4288414141 hasConceptScore W4288414141C86803240 @default.
- W4288414141 hasConceptScore W4288414141C97355855 @default.
- W4288414141 hasConceptScore W4288414141C97541855 @default.
- W4288414141 hasLocation W42884141411 @default.
- W4288414141 hasOpenAccess W4288414141 @default.
- W4288414141 hasPrimaryLocation W42884141411 @default.
- W4288414141 hasRelatedWork W1494614182 @default.
- W4288414141 hasRelatedWork W1997664188 @default.
- W4288414141 hasRelatedWork W2960456850 @default.
- W4288414141 hasRelatedWork W3022038857 @default.
- W4288414141 hasRelatedWork W3131673289 @default.
- W4288414141 hasRelatedWork W4281382123 @default.
- W4288414141 hasRelatedWork W4281645081 @default.
- W4288414141 hasRelatedWork W4308262314 @default.
- W4288414141 hasRelatedWork W4318834068 @default.
- W4288414141 hasRelatedWork W4319083788 @default.
- W4288414141 isParatext "false" @default.
- W4288414141 isRetracted "false" @default.
- W4288414141 workType "article" @default.