Matches in SemOpenAlex for { <https://semopenalex.org/work/W4224256292> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4224256292 abstract "Model-based reinforcement learning (RL) algorithms designed for handling complex visual observations typically learn some sort of latent state representation, either explicitly or implicitly. Standard methods of this sort do not distinguish between functionally relevant aspects of the state and irrelevant distractors, instead aiming to represent all available information equally. We propose a modified objective for model-based RL that, in combination with mutual information maximization, allows us to learn representations and dynamics for visual model-based RL without reconstruction in a way that explicitly prioritizes functionally relevant factors. The key principle behind our design is to integrate a term inspired by variational empowerment into a state-space model based on mutual information. This term prioritizes information that is correlated with action, thus ensuring that functionally relevant factors are captured first. Furthermore, the same empowerment term also promotes faster exploration during the RL process, especially for sparse-reward tasks where the reward signal is insufficient to drive exploration in the early stages of learning. We evaluate the approach on a suite of vision-based robot control tasks with natural video backgrounds, and show that the proposed prioritized information objective outperforms state-of-the-art model based RL approaches with higher sample efficiency and episodic returns. https://sites.google.com/view/information-empowerment" @default.
- W4224256292 created "2022-04-26" @default.
- W4224256292 creator A5026322200 @default.
- W4224256292 creator A5049277598 @default.
- W4224256292 creator A5064794691 @default.
- W4224256292 creator A5084125668 @default.
- W4224256292 date "2022-04-18" @default.
- W4224256292 modified "2023-09-26" @default.
- W4224256292 title "INFOrmation Prioritization through EmPOWERment in Visual Model-Based RL" @default.
- W4224256292 doi "https://doi.org/10.48550/arxiv.2204.08585" @default.
- W4224256292 hasPublicationYear "2022" @default.
- W4224256292 type Work @default.
- W4224256292 citedByCount "0" @default.
- W4224256292 crossrefType "posted-content" @default.
- W4224256292 hasAuthorship W4224256292A5026322200 @default.
- W4224256292 hasAuthorship W4224256292A5049277598 @default.
- W4224256292 hasAuthorship W4224256292A5064794691 @default.
- W4224256292 hasAuthorship W4224256292A5084125668 @default.
- W4224256292 hasBestOaLocation W42242562921 @default.
- W4224256292 hasConcept C111919701 @default.
- W4224256292 hasConcept C119857082 @default.
- W4224256292 hasConcept C126255220 @default.
- W4224256292 hasConcept C152139883 @default.
- W4224256292 hasConcept C154945302 @default.
- W4224256292 hasConcept C17744445 @default.
- W4224256292 hasConcept C199539241 @default.
- W4224256292 hasConcept C20555606 @default.
- W4224256292 hasConcept C23123220 @default.
- W4224256292 hasConcept C2776330181 @default.
- W4224256292 hasConcept C2776359362 @default.
- W4224256292 hasConcept C33923547 @default.
- W4224256292 hasConcept C41008148 @default.
- W4224256292 hasConcept C88548561 @default.
- W4224256292 hasConcept C90509273 @default.
- W4224256292 hasConcept C94625758 @default.
- W4224256292 hasConcept C97541855 @default.
- W4224256292 hasConcept C98045186 @default.
- W4224256292 hasConceptScore W4224256292C111919701 @default.
- W4224256292 hasConceptScore W4224256292C119857082 @default.
- W4224256292 hasConceptScore W4224256292C126255220 @default.
- W4224256292 hasConceptScore W4224256292C152139883 @default.
- W4224256292 hasConceptScore W4224256292C154945302 @default.
- W4224256292 hasConceptScore W4224256292C17744445 @default.
- W4224256292 hasConceptScore W4224256292C199539241 @default.
- W4224256292 hasConceptScore W4224256292C20555606 @default.
- W4224256292 hasConceptScore W4224256292C23123220 @default.
- W4224256292 hasConceptScore W4224256292C2776330181 @default.
- W4224256292 hasConceptScore W4224256292C2776359362 @default.
- W4224256292 hasConceptScore W4224256292C33923547 @default.
- W4224256292 hasConceptScore W4224256292C41008148 @default.
- W4224256292 hasConceptScore W4224256292C88548561 @default.
- W4224256292 hasConceptScore W4224256292C90509273 @default.
- W4224256292 hasConceptScore W4224256292C94625758 @default.
- W4224256292 hasConceptScore W4224256292C97541855 @default.
- W4224256292 hasConceptScore W4224256292C98045186 @default.
- W4224256292 hasLocation W42242562921 @default.
- W4224256292 hasOpenAccess W4224256292 @default.
- W4224256292 hasPrimaryLocation W42242562921 @default.
- W4224256292 hasRelatedWork W2897101624 @default.
- W4224256292 hasRelatedWork W2993842823 @default.
- W4224256292 hasRelatedWork W3022038857 @default.
- W4224256292 hasRelatedWork W3088447613 @default.
- W4224256292 hasRelatedWork W3106587939 @default.
- W4224256292 hasRelatedWork W4210841218 @default.
- W4224256292 hasRelatedWork W4224951294 @default.
- W4224256292 hasRelatedWork W4287997019 @default.
- W4224256292 hasRelatedWork W4319083788 @default.
- W4224256292 hasRelatedWork W4375870063 @default.
- W4224256292 isParatext "false" @default.
- W4224256292 isRetracted "false" @default.
- W4224256292 workType "article" @default.