Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312772252> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4312772252 abstract "Learning from demonstration (LfD) techniques seek to enable users without computer programming experience to teach robots novel tasks. There are generally two types of LfD: human- and robot-centric. While human-centric learning is intuitive, human centric learning suffers from performance degradation due to covariate shift. Robot-centric approaches, such as Dataset Aggregation (DAgger), address covariate shift but can struggle to learn from suboptimal human teachers. To create a more human-aware version of robot-centric LfD, we present Mutual Information-driven Meta-learning from Demonstration (MIND MELD). MIND MELD meta-learns a mapping from suboptimal and heterogeneous human feedback to optimal labels, thereby improving the learning signal for robot-centric LfD. The key to our approach is learning an informative personalized em-bedding using mutual information maximization via variational inference. The embedding then informs a mapping from human provided labels to optimal labels. We evaluate our framework in a human-subjects experiment, demonstrating that our approach improves corrective labels provided by human demonstrators. Our framework outperforms baselines in terms of ability to reach the goal <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$(p <. 001)$</tex> , average distance from the goal <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$(p=.006)$</tex> , and various subjective ratings <tex xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>$(p=.008)$</tex> ." @default.
- W4312772252 created "2023-01-05" @default.
- W4312772252 creator A5001734584 @default.
- W4312772252 creator A5008211323 @default.
- W4312772252 creator A5020550513 @default.
- W4312772252 creator A5046734455 @default.
- W4312772252 date "2022-03-07" @default.
- W4312772252 modified "2023-10-05" @default.
- W4312772252 title "MIND MELD: Personalized Meta-Learning for Robot-Centric Imitation Learning" @default.
- W4312772252 cites W1501005121 @default.
- W4312772252 cites W1980969546 @default.
- W4312772252 cites W1986014385 @default.
- W4312772252 cites W2032568497 @default.
- W4312772252 cites W2044140042 @default.
- W4312772252 cites W2053910308 @default.
- W4312772252 cites W2110064869 @default.
- W4312772252 cites W2131799829 @default.
- W4312772252 cites W2154018708 @default.
- W4312772252 cites W2157289187 @default.
- W4312772252 cites W2409942531 @default.
- W4312772252 cites W2527925052 @default.
- W4312772252 cites W2903139669 @default.
- W4312772252 cites W3003342008 @default.
- W4312772252 cites W3014235025 @default.
- W4312772252 doi "https://doi.org/10.1109/hri53351.2022.9889616" @default.
- W4312772252 hasPublicationYear "2022" @default.
- W4312772252 type Work @default.
- W4312772252 citedByCount "2" @default.
- W4312772252 countsByYear W43127722522022 @default.
- W4312772252 countsByYear W43127722522023 @default.
- W4312772252 crossrefType "proceedings-article" @default.
- W4312772252 hasAuthorship W4312772252A5001734584 @default.
- W4312772252 hasAuthorship W4312772252A5008211323 @default.
- W4312772252 hasAuthorship W4312772252A5020550513 @default.
- W4312772252 hasAuthorship W4312772252A5046734455 @default.
- W4312772252 hasConcept C107457646 @default.
- W4312772252 hasConcept C119857082 @default.
- W4312772252 hasConcept C126255220 @default.
- W4312772252 hasConcept C154945302 @default.
- W4312772252 hasConcept C26517878 @default.
- W4312772252 hasConcept C2776214188 @default.
- W4312772252 hasConcept C2776330181 @default.
- W4312772252 hasConcept C33923547 @default.
- W4312772252 hasConcept C38652104 @default.
- W4312772252 hasConcept C41008148 @default.
- W4312772252 hasConcept C90509273 @default.
- W4312772252 hasConceptScore W4312772252C107457646 @default.
- W4312772252 hasConceptScore W4312772252C119857082 @default.
- W4312772252 hasConceptScore W4312772252C126255220 @default.
- W4312772252 hasConceptScore W4312772252C154945302 @default.
- W4312772252 hasConceptScore W4312772252C26517878 @default.
- W4312772252 hasConceptScore W4312772252C2776214188 @default.
- W4312772252 hasConceptScore W4312772252C2776330181 @default.
- W4312772252 hasConceptScore W4312772252C33923547 @default.
- W4312772252 hasConceptScore W4312772252C38652104 @default.
- W4312772252 hasConceptScore W4312772252C41008148 @default.
- W4312772252 hasConceptScore W4312772252C90509273 @default.
- W4312772252 hasLocation W43127722521 @default.
- W4312772252 hasOpenAccess W4312772252 @default.
- W4312772252 hasPrimaryLocation W43127722521 @default.
- W4312772252 hasRelatedWork W2329452785 @default.
- W4312772252 hasRelatedWork W2356380379 @default.
- W4312772252 hasRelatedWork W2961085424 @default.
- W4312772252 hasRelatedWork W3046775127 @default.
- W4312772252 hasRelatedWork W3170094116 @default.
- W4312772252 hasRelatedWork W4285260836 @default.
- W4312772252 hasRelatedWork W4286629047 @default.
- W4312772252 hasRelatedWork W4306321456 @default.
- W4312772252 hasRelatedWork W4306674287 @default.
- W4312772252 hasRelatedWork W4224009465 @default.
- W4312772252 isParatext "false" @default.
- W4312772252 isRetracted "false" @default.
- W4312772252 workType "article" @default.