Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387636046> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4387636046 abstract "Unified models capable of solving a wide variety of tasks have gained traction in vision and NLP due to their ability to share regularities and structures across tasks, which improves individual task performance and reduces computational footprint. However, the impact of such models remains limited in embodied learning problems, which present unique challenges due to interactivity, sample inefficiency, and sequential task presentation. In this work, we present PolyTask, a novel method for learning a single unified model that can solve various embodied tasks through a 'learn then distill' mechanism. In the 'learn' step, PolyTask leverages a few demonstrations for each task to train task-specific policies. Then, in the 'distill' step, task-specific policies are distilled into a single policy using a new distillation method called Behavior Distillation. Given a unified policy, individual task behavior can be extracted through conditioning variables. PolyTask is designed to be conceptually simple while being able to leverage well-established algorithms in RL to enable interactivity, a handful of expert demonstrations to allow for sample efficiency, and preventing interactive access to tasks during distillation to enable lifelong learning. Experiments across three simulated environment suites and a real-robot suite show that PolyTask outperforms prior state-of-the-art approaches in multi-task and lifelong learning settings by significant margins." @default.
- W4387636046 created "2023-10-14" @default.
- W4387636046 creator A5035065997 @default.
- W4387636046 creator A5056646668 @default.
- W4387636046 date "2023-10-12" @default.
- W4387636046 modified "2023-10-15" @default.
- W4387636046 title "PolyTask: Learning Unified Policies through Behavior Distillation" @default.
- W4387636046 doi "https://doi.org/10.48550/arxiv.2310.08573" @default.
- W4387636046 hasPublicationYear "2023" @default.
- W4387636046 type Work @default.
- W4387636046 citedByCount "0" @default.
- W4387636046 crossrefType "posted-content" @default.
- W4387636046 hasAuthorship W4387636046A5035065997 @default.
- W4387636046 hasAuthorship W4387636046A5056646668 @default.
- W4387636046 hasBestOaLocation W43876360461 @default.
- W4387636046 hasConcept C107457646 @default.
- W4387636046 hasConcept C119857082 @default.
- W4387636046 hasConcept C127413603 @default.
- W4387636046 hasConcept C144430266 @default.
- W4387636046 hasConcept C153083717 @default.
- W4387636046 hasConcept C154945302 @default.
- W4387636046 hasConcept C178790620 @default.
- W4387636046 hasConcept C185592680 @default.
- W4387636046 hasConcept C201995342 @default.
- W4387636046 hasConcept C204030448 @default.
- W4387636046 hasConcept C2776999362 @default.
- W4387636046 hasConcept C2780451532 @default.
- W4387636046 hasConcept C41008148 @default.
- W4387636046 hasConcept C49774154 @default.
- W4387636046 hasConcept C97541855 @default.
- W4387636046 hasConceptScore W4387636046C107457646 @default.
- W4387636046 hasConceptScore W4387636046C119857082 @default.
- W4387636046 hasConceptScore W4387636046C127413603 @default.
- W4387636046 hasConceptScore W4387636046C144430266 @default.
- W4387636046 hasConceptScore W4387636046C153083717 @default.
- W4387636046 hasConceptScore W4387636046C154945302 @default.
- W4387636046 hasConceptScore W4387636046C178790620 @default.
- W4387636046 hasConceptScore W4387636046C185592680 @default.
- W4387636046 hasConceptScore W4387636046C201995342 @default.
- W4387636046 hasConceptScore W4387636046C204030448 @default.
- W4387636046 hasConceptScore W4387636046C2776999362 @default.
- W4387636046 hasConceptScore W4387636046C2780451532 @default.
- W4387636046 hasConceptScore W4387636046C41008148 @default.
- W4387636046 hasConceptScore W4387636046C49774154 @default.
- W4387636046 hasConceptScore W4387636046C97541855 @default.
- W4387636046 hasLocation W43876360461 @default.
- W4387636046 hasOpenAccess W4387636046 @default.
- W4387636046 hasPrimaryLocation W43876360461 @default.
- W4387636046 hasRelatedWork W1513662110 @default.
- W4387636046 hasRelatedWork W1975913006 @default.
- W4387636046 hasRelatedWork W2013411520 @default.
- W4387636046 hasRelatedWork W2063506784 @default.
- W4387636046 hasRelatedWork W2272707781 @default.
- W4387636046 hasRelatedWork W2319020389 @default.
- W4387636046 hasRelatedWork W2521348551 @default.
- W4387636046 hasRelatedWork W2747851897 @default.
- W4387636046 hasRelatedWork W2768698792 @default.
- W4387636046 hasRelatedWork W4240086805 @default.
- W4387636046 isParatext "false" @default.
- W4387636046 isRetracted "false" @default.
- W4387636046 workType "article" @default.