Matches in SemOpenAlex for { <https://semopenalex.org/work/W3003881683> ?p ?o ?g. }
- W3003881683 abstract "We study self-supervised adaptation of a robot's policy for social interaction, i.e., a policy for active communication with surrounding pedestrians through audio or visual signals. Inspired by the observation that humans continually adapt their behavior when interacting under varying social context, we propose Adaptive EXP4 (A-EXP4), a novel online learning algorithm for adapting the robot-pedestrian interaction policy. To address limitations of bandit algorithms in adaptation to unseen and highly dynamic scenarios, we employ a mixture model over the policy parameter space. Specifically, a Dirichlet Process Gaussian Mixture Model (DPMM) is used to cluster the parameters of sampled policies and maintain a mixture model over the clusters, hence effectively discovering policies that are suitable to the current environmental context in an unsupervised manner. Our simulated and real-world experiments demonstrate the feasibility of A-EXP4 in accommodating interaction with different types of pedestrians while jointly minimizing social disruption through the adaptation process. While the A-EXP4 formulation is kept general for application in a variety of domains requiring continual adaptation of a robot's policy, we specifically evaluate the performance of our algorithm using a suitcase-inspired assistive robotic platform. In this concrete assistive scenario, the algorithm observes how audio signals produced by the navigational system affect the behavior of pedestrians and adapts accordingly. Consequently, we find A-EXP4 to effectively adapt the interaction policy for gently clearing a navigation path in crowded settings, resulting in significant reduction in empirical regret compared to the EXP4 baseline." @default.
- W3003881683 created "2020-02-07" @default.
- W3003881683 creator A5025893246 @default.
- W3003881683 creator A5037322163 @default.
- W3003881683 creator A5088881074 @default.
- W3003881683 creator A5091550931 @default.
- W3003881683 date "2019-11-01" @default.
- W3003881683 modified "2023-09-26" @default.
- W3003881683 title "A-EXP4: Online Social Policy Learning for Adaptive Robot-Pedestrian Interaction" @default.
- W3003881683 cites W123233698 @default.
- W3003881683 cites W1579979603 @default.
- W3003881683 cites W1651456183 @default.
- W3003881683 cites W1912700219 @default.
- W3003881683 cites W1969173200 @default.
- W3003881683 cites W1977655452 @default.
- W3003881683 cites W2020999234 @default.
- W3003881683 cites W2029780883 @default.
- W3003881683 cites W2077902449 @default.
- W3003881683 cites W2088861238 @default.
- W3003881683 cites W2101821104 @default.
- W3003881683 cites W2103715332 @default.
- W3003881683 cites W2105934661 @default.
- W3003881683 cites W2119717200 @default.
- W3003881683 cites W2120070743 @default.
- W3003881683 cites W2127107099 @default.
- W3003881683 cites W2127498532 @default.
- W3003881683 cites W2137104525 @default.
- W3003881683 cites W2140135625 @default.
- W3003881683 cites W2160609653 @default.
- W3003881683 cites W2162932021 @default.
- W3003881683 cites W2167117957 @default.
- W3003881683 cites W2173248099 @default.
- W3003881683 cites W2256838191 @default.
- W3003881683 cites W2341852456 @default.
- W3003881683 cites W2343174257 @default.
- W3003881683 cites W2424778531 @default.
- W3003881683 cites W2463627759 @default.
- W3003881683 cites W2512654307 @default.
- W3003881683 cites W2536057734 @default.
- W3003881683 cites W2537547826 @default.
- W3003881683 cites W2565370028 @default.
- W3003881683 cites W2736730521 @default.
- W3003881683 cites W2765252989 @default.
- W3003881683 cites W2817973999 @default.
- W3003881683 cites W2888465049 @default.
- W3003881683 cites W2889628315 @default.
- W3003881683 cites W2911273949 @default.
- W3003881683 cites W2963888093 @default.
- W3003881683 cites W2964161785 @default.
- W3003881683 cites W2964338736 @default.
- W3003881683 cites W657026121 @default.
- W3003881683 cites W77456527 @default.
- W3003881683 doi "https://doi.org/10.1109/iros40897.2019.8967737" @default.
- W3003881683 hasPublicationYear "2019" @default.
- W3003881683 type Work @default.
- W3003881683 sameAs 3003881683 @default.
- W3003881683 citedByCount "0" @default.
- W3003881683 crossrefType "proceedings-article" @default.
- W3003881683 hasAuthorship W3003881683A5025893246 @default.
- W3003881683 hasAuthorship W3003881683A5037322163 @default.
- W3003881683 hasAuthorship W3003881683A5088881074 @default.
- W3003881683 hasAuthorship W3003881683A5091550931 @default.
- W3003881683 hasConcept C107457646 @default.
- W3003881683 hasConcept C111919701 @default.
- W3003881683 hasConcept C119857082 @default.
- W3003881683 hasConcept C120665830 @default.
- W3003881683 hasConcept C121332964 @default.
- W3003881683 hasConcept C127413603 @default.
- W3003881683 hasConcept C139807058 @default.
- W3003881683 hasConcept C151730666 @default.
- W3003881683 hasConcept C154945302 @default.
- W3003881683 hasConcept C163716315 @default.
- W3003881683 hasConcept C22212356 @default.
- W3003881683 hasConcept C2777113093 @default.
- W3003881683 hasConcept C2779343474 @default.
- W3003881683 hasConcept C41008148 @default.
- W3003881683 hasConcept C50817715 @default.
- W3003881683 hasConcept C61224824 @default.
- W3003881683 hasConcept C61326573 @default.
- W3003881683 hasConcept C62520636 @default.
- W3003881683 hasConcept C86803240 @default.
- W3003881683 hasConcept C90509273 @default.
- W3003881683 hasConcept C97541855 @default.
- W3003881683 hasConcept C98045186 @default.
- W3003881683 hasConceptScore W3003881683C107457646 @default.
- W3003881683 hasConceptScore W3003881683C111919701 @default.
- W3003881683 hasConceptScore W3003881683C119857082 @default.
- W3003881683 hasConceptScore W3003881683C120665830 @default.
- W3003881683 hasConceptScore W3003881683C121332964 @default.
- W3003881683 hasConceptScore W3003881683C127413603 @default.
- W3003881683 hasConceptScore W3003881683C139807058 @default.
- W3003881683 hasConceptScore W3003881683C151730666 @default.
- W3003881683 hasConceptScore W3003881683C154945302 @default.
- W3003881683 hasConceptScore W3003881683C163716315 @default.
- W3003881683 hasConceptScore W3003881683C22212356 @default.
- W3003881683 hasConceptScore W3003881683C2777113093 @default.
- W3003881683 hasConceptScore W3003881683C2779343474 @default.
- W3003881683 hasConceptScore W3003881683C41008148 @default.
- W3003881683 hasConceptScore W3003881683C50817715 @default.
- W3003881683 hasConceptScore W3003881683C61224824 @default.