Matches in SemOpenAlex for { <https://semopenalex.org/work/W4363671520> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W4363671520 abstract "Approaches for teaching learning agents via human demonstrations have been widely studied and successfully applied to multiple domains. However, the majority of imitation learning work utilizes only behavioral information from the demonstrator, i.e. which actions were taken, and ignores other useful information. In particular, eye gaze information can give valuable insight towards where the demonstrator is allocating visual attention, and holds the potential to improve agent performance and generalization. In this work, we propose Gaze Regularized Imitation Learning (GRIL), a novel context-aware, imitation learning architecture that learns concurrently from both human demonstrations and eye gaze to solve tasks where visual attention provides important context. We apply GRIL to a visual navigation task, in which an unmanned quadrotor is trained to search for and navigate to a target vehicle in a photorealistic simulated environment. We show that GRIL outperforms several state-of-the-art gaze-based imitation learning algorithms, simultaneously learns to predict human visual attention, and generalizes to scenarios not present in the training data. Supplemental videos and code can be found at https://sites.google.com/view/gaze-regularized-il/." @default.
- W4363671520 created "2023-04-11" @default.
- W4363671520 creator A5007706088 @default.
- W4363671520 creator A5010917268 @default.
- W4363671520 creator A5040988341 @default.
- W4363671520 creator A5042275001 @default.
- W4363671520 creator A5051690828 @default.
- W4363671520 creator A5057151264 @default.
- W4363671520 creator A5065387163 @default.
- W4363671520 creator A5069949825 @default.
- W4363671520 creator A5079459509 @default.
- W4363671520 date "2021-02-25" @default.
- W4363671520 modified "2023-10-16" @default.
- W4363671520 title "Imitation Learning with Human Eye Gaze via Multi-Objective Prediction" @default.
- W4363671520 doi "https://doi.org/10.48550/arxiv.2102.13008" @default.
- W4363671520 hasPublicationYear "2021" @default.
- W4363671520 type Work @default.
- W4363671520 citedByCount "0" @default.
- W4363671520 crossrefType "posted-content" @default.
- W4363671520 hasAuthorship W4363671520A5007706088 @default.
- W4363671520 hasAuthorship W4363671520A5010917268 @default.
- W4363671520 hasAuthorship W4363671520A5040988341 @default.
- W4363671520 hasAuthorship W4363671520A5042275001 @default.
- W4363671520 hasAuthorship W4363671520A5051690828 @default.
- W4363671520 hasAuthorship W4363671520A5057151264 @default.
- W4363671520 hasAuthorship W4363671520A5065387163 @default.
- W4363671520 hasAuthorship W4363671520A5069949825 @default.
- W4363671520 hasAuthorship W4363671520A5079459509 @default.
- W4363671520 hasBestOaLocation W43636715201 @default.
- W4363671520 hasConcept C107457646 @default.
- W4363671520 hasConcept C126388530 @default.
- W4363671520 hasConcept C134306372 @default.
- W4363671520 hasConcept C151730666 @default.
- W4363671520 hasConcept C154945302 @default.
- W4363671520 hasConcept C15744967 @default.
- W4363671520 hasConcept C158495155 @default.
- W4363671520 hasConcept C162324750 @default.
- W4363671520 hasConcept C177148314 @default.
- W4363671520 hasConcept C180747234 @default.
- W4363671520 hasConcept C187736073 @default.
- W4363671520 hasConcept C2779321571 @default.
- W4363671520 hasConcept C2779343474 @default.
- W4363671520 hasConcept C2779916870 @default.
- W4363671520 hasConcept C2780451532 @default.
- W4363671520 hasConcept C31972630 @default.
- W4363671520 hasConcept C33923547 @default.
- W4363671520 hasConcept C41008148 @default.
- W4363671520 hasConcept C56461940 @default.
- W4363671520 hasConcept C77805123 @default.
- W4363671520 hasConcept C86803240 @default.
- W4363671520 hasConceptScore W4363671520C107457646 @default.
- W4363671520 hasConceptScore W4363671520C126388530 @default.
- W4363671520 hasConceptScore W4363671520C134306372 @default.
- W4363671520 hasConceptScore W4363671520C151730666 @default.
- W4363671520 hasConceptScore W4363671520C154945302 @default.
- W4363671520 hasConceptScore W4363671520C15744967 @default.
- W4363671520 hasConceptScore W4363671520C158495155 @default.
- W4363671520 hasConceptScore W4363671520C162324750 @default.
- W4363671520 hasConceptScore W4363671520C177148314 @default.
- W4363671520 hasConceptScore W4363671520C180747234 @default.
- W4363671520 hasConceptScore W4363671520C187736073 @default.
- W4363671520 hasConceptScore W4363671520C2779321571 @default.
- W4363671520 hasConceptScore W4363671520C2779343474 @default.
- W4363671520 hasConceptScore W4363671520C2779916870 @default.
- W4363671520 hasConceptScore W4363671520C2780451532 @default.
- W4363671520 hasConceptScore W4363671520C31972630 @default.
- W4363671520 hasConceptScore W4363671520C33923547 @default.
- W4363671520 hasConceptScore W4363671520C41008148 @default.
- W4363671520 hasConceptScore W4363671520C56461940 @default.
- W4363671520 hasConceptScore W4363671520C77805123 @default.
- W4363671520 hasConceptScore W4363671520C86803240 @default.
- W4363671520 hasLocation W43636715201 @default.
- W4363671520 hasOpenAccess W4363671520 @default.
- W4363671520 hasPrimaryLocation W43636715201 @default.
- W4363671520 hasRelatedWork W1561602432 @default.
- W4363671520 hasRelatedWork W1589443630 @default.
- W4363671520 hasRelatedWork W1999328396 @default.
- W4363671520 hasRelatedWork W2127677160 @default.
- W4363671520 hasRelatedWork W2148584227 @default.
- W4363671520 hasRelatedWork W2168991662 @default.
- W4363671520 hasRelatedWork W2385108104 @default.
- W4363671520 hasRelatedWork W2769824675 @default.
- W4363671520 hasRelatedWork W3033353262 @default.
- W4363671520 hasRelatedWork W2183271128 @default.
- W4363671520 isParatext "false" @default.
- W4363671520 isRetracted "false" @default.
- W4363671520 workType "article" @default.