Matches in SemOpenAlex for { <https://semopenalex.org/work/W3089341726> ?p ?o ?g. }
- W3089341726 abstract "Planners using accurate models can be effective for accomplishing manipulation tasks in the real world, but are typically highly specialized and require significant fine-tuning to be reliable. Meanwhile, learning is useful for adaptation, but can require a substantial amount of data collection. In this paper, we propose a method that improves the efficiency of sub-optimal planners with approximate but simple and fast models by switching to a model-free policy when unexpected transitions are observed. Unlike previous work, our method specifically addresses when the planner fails due to transition model error by patching with a local policy only where needed. First, we use a sub-optimal model-based planner to perform a task until model failure is detected. Next, we learn a local model-free policy from expert demonstrations to complete the task in regions where the model failed. To show the efficacy of our method, we perform experiments with a shape insertion puzzle and compare our results to both pure planning and imitation learning approaches. We then apply our method to a door opening task. Our experiments demonstrate that our patch-enhanced planner performs more reliably than pure planning and with lower overall sample complexity than pure imitation learning." @default.
- W3089341726 created "2020-10-08" @default.
- W3089341726 creator A5005091065 @default.
- W3089341726 creator A5021424850 @default.
- W3089341726 creator A5032956379 @default.
- W3089341726 date "2020-09-29" @default.
- W3089341726 modified "2023-09-26" @default.
- W3089341726 title "Learning Skills to Patch Plans Based on Inaccurate Models" @default.
- W3089341726 cites W131069610 @default.
- W3089341726 cites W1746819321 @default.
- W3089341726 cites W1952489873 @default.
- W3089341726 cites W2005126631 @default.
- W3089341726 cites W2025998811 @default.
- W3089341726 cites W2029342956 @default.
- W3089341726 cites W2042408133 @default.
- W3089341726 cites W2101234009 @default.
- W3089341726 cites W2101340954 @default.
- W3089341726 cites W2108535023 @default.
- W3089341726 cites W2109910161 @default.
- W3089341726 cites W2121103318 @default.
- W3089341726 cites W2474696542 @default.
- W3089341726 cites W2526512233 @default.
- W3089341726 cites W2565233142 @default.
- W3089341726 cites W2772449134 @default.
- W3089341726 cites W2774527530 @default.
- W3089341726 cites W2794666284 @default.
- W3089341726 cites W2803281228 @default.
- W3089341726 cites W2805883505 @default.
- W3089341726 cites W2898585858 @default.
- W3089341726 cites W2909491791 @default.
- W3089341726 cites W2925883569 @default.
- W3089341726 cites W2951004968 @default.
- W3089341726 cites W2951458352 @default.
- W3089341726 cites W2959488596 @default.
- W3089341726 cites W2962732055 @default.
- W3089341726 cites W2963166639 @default.
- W3089341726 cites W2963411833 @default.
- W3089341726 cites W2963634205 @default.
- W3089341726 cites W2964091363 @default.
- W3089341726 cites W2968095426 @default.
- W3089341726 cites W2980473643 @default.
- W3089341726 cites W2990747716 @default.
- W3089341726 cites W2991934429 @default.
- W3089341726 cites W3040464380 @default.
- W3089341726 cites W3091552533 @default.
- W3089341726 hasPublicationYear "2020" @default.
- W3089341726 type Work @default.
- W3089341726 sameAs 3089341726 @default.
- W3089341726 citedByCount "0" @default.
- W3089341726 crossrefType "posted-content" @default.
- W3089341726 hasAuthorship W3089341726A5005091065 @default.
- W3089341726 hasAuthorship W3089341726A5021424850 @default.
- W3089341726 hasAuthorship W3089341726A5032956379 @default.
- W3089341726 hasConcept C111472728 @default.
- W3089341726 hasConcept C119857082 @default.
- W3089341726 hasConcept C120665830 @default.
- W3089341726 hasConcept C121332964 @default.
- W3089341726 hasConcept C126388530 @default.
- W3089341726 hasConcept C127413603 @default.
- W3089341726 hasConcept C138885662 @default.
- W3089341726 hasConcept C139807058 @default.
- W3089341726 hasConcept C154945302 @default.
- W3089341726 hasConcept C15744967 @default.
- W3089341726 hasConcept C18762648 @default.
- W3089341726 hasConcept C201995342 @default.
- W3089341726 hasConcept C2776999362 @default.
- W3089341726 hasConcept C2779436431 @default.
- W3089341726 hasConcept C2780451532 @default.
- W3089341726 hasConcept C2780586882 @default.
- W3089341726 hasConcept C41008148 @default.
- W3089341726 hasConcept C77805123 @default.
- W3089341726 hasConcept C78519656 @default.
- W3089341726 hasConceptScore W3089341726C111472728 @default.
- W3089341726 hasConceptScore W3089341726C119857082 @default.
- W3089341726 hasConceptScore W3089341726C120665830 @default.
- W3089341726 hasConceptScore W3089341726C121332964 @default.
- W3089341726 hasConceptScore W3089341726C126388530 @default.
- W3089341726 hasConceptScore W3089341726C127413603 @default.
- W3089341726 hasConceptScore W3089341726C138885662 @default.
- W3089341726 hasConceptScore W3089341726C139807058 @default.
- W3089341726 hasConceptScore W3089341726C154945302 @default.
- W3089341726 hasConceptScore W3089341726C15744967 @default.
- W3089341726 hasConceptScore W3089341726C18762648 @default.
- W3089341726 hasConceptScore W3089341726C201995342 @default.
- W3089341726 hasConceptScore W3089341726C2776999362 @default.
- W3089341726 hasConceptScore W3089341726C2779436431 @default.
- W3089341726 hasConceptScore W3089341726C2780451532 @default.
- W3089341726 hasConceptScore W3089341726C2780586882 @default.
- W3089341726 hasConceptScore W3089341726C41008148 @default.
- W3089341726 hasConceptScore W3089341726C77805123 @default.
- W3089341726 hasConceptScore W3089341726C78519656 @default.
- W3089341726 hasOpenAccess W3089341726 @default.
- W3089341726 hasRelatedWork W2295763688 @default.
- W3089341726 hasRelatedWork W2613603362 @default.
- W3089341726 hasRelatedWork W2626860042 @default.
- W3089341726 hasRelatedWork W2770014065 @default.
- W3089341726 hasRelatedWork W2803281228 @default.
- W3089341726 hasRelatedWork W2951456741 @default.
- W3089341726 hasRelatedWork W2979605429 @default.
- W3089341726 hasRelatedWork W2981547136 @default.