Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297688052> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4297688052 endingPage "1878" @default.
- W4297688052 startingPage "1866" @default.
- W4297688052 abstract "Abstract Visual navigation task is to steer an embodied agent finding the given target based on observation. The effective transformer from observation of the agent to visual representation determines the navigation actions and promotes more informed navigation policy. In this work, we propose a spatial sequential transformer network (SSTNet) for learning informative visual representation in deep reinforcement learning. SSTNet is composed by spatial attention probability fused model (SAF) and sequential transformer network (STNet). SAF enforces cross-modal state into visual clues in reinforcement learning. It encodes semantic information about observed objects, as well as spatial information about their location, which jointly exploiting image inter-relations. STNet generates (imagines) the next observations and makes action inference of the aspects most relevant to the target. It decodes the image intra-relations. This way, the agent learns to understand the causality between navigation actions and dynamic changes in observations. SSTNet is conditioned on an auto-regressive model on the desired reward, past states, actions, and knowledge graph. The whole navigation framework considers the local and global visual information, as well as time sequential information. Thus, it allows the agent to navigate towards the sought-after object effectively. We evaluate our model on the AI2THOR framework show that our method attains at least $10%$ improvement of average success rate over most state-of-the-art models. Code and datasets can be found in https://github.com/zhoukang123/SDTNet_2022." @default.
- W4297688052 created "2022-09-30" @default.
- W4297688052 creator A5005228021 @default.
- W4297688052 creator A5053880803 @default.
- W4297688052 creator A5068416093 @default.
- W4297688052 date "2022-08-31" @default.
- W4297688052 modified "2023-10-18" @default.
- W4297688052 title "TransNav: spatial sequential transformer network for visual navigation" @default.
- W4297688052 cites W2250539671 @default.
- W4297688052 cites W2277195237 @default.
- W4297688052 cites W2800142021 @default.
- W4297688052 cites W2962887844 @default.
- W4297688052 cites W2965964939 @default.
- W4297688052 cites W2969277962 @default.
- W4297688052 cites W2984100107 @default.
- W4297688052 cites W3035869396 @default.
- W4297688052 cites W3107094551 @default.
- W4297688052 cites W4236558809 @default.
- W4297688052 cites W639708223 @default.
- W4297688052 doi "https://doi.org/10.1093/jcde/qwac084" @default.
- W4297688052 hasPublicationYear "2022" @default.
- W4297688052 type Work @default.
- W4297688052 citedByCount "1" @default.
- W4297688052 countsByYear W42976880522023 @default.
- W4297688052 crossrefType "journal-article" @default.
- W4297688052 hasAuthorship W4297688052A5005228021 @default.
- W4297688052 hasAuthorship W4297688052A5053880803 @default.
- W4297688052 hasAuthorship W4297688052A5068416093 @default.
- W4297688052 hasBestOaLocation W42976880521 @default.
- W4297688052 hasConcept C119599485 @default.
- W4297688052 hasConcept C119857082 @default.
- W4297688052 hasConcept C127413603 @default.
- W4297688052 hasConcept C154945302 @default.
- W4297688052 hasConcept C165801399 @default.
- W4297688052 hasConcept C2776214188 @default.
- W4297688052 hasConcept C41008148 @default.
- W4297688052 hasConcept C66322947 @default.
- W4297688052 hasConcept C97541855 @default.
- W4297688052 hasConceptScore W4297688052C119599485 @default.
- W4297688052 hasConceptScore W4297688052C119857082 @default.
- W4297688052 hasConceptScore W4297688052C127413603 @default.
- W4297688052 hasConceptScore W4297688052C154945302 @default.
- W4297688052 hasConceptScore W4297688052C165801399 @default.
- W4297688052 hasConceptScore W4297688052C2776214188 @default.
- W4297688052 hasConceptScore W4297688052C41008148 @default.
- W4297688052 hasConceptScore W4297688052C66322947 @default.
- W4297688052 hasConceptScore W4297688052C97541855 @default.
- W4297688052 hasFunder F4320310282 @default.
- W4297688052 hasFunder F4320335777 @default.
- W4297688052 hasIssue "5" @default.
- W4297688052 hasLocation W42976880521 @default.
- W4297688052 hasOpenAccess W4297688052 @default.
- W4297688052 hasPrimaryLocation W42976880521 @default.
- W4297688052 hasRelatedWork W260766989 @default.
- W4297688052 hasRelatedWork W2959276766 @default.
- W4297688052 hasRelatedWork W2961085424 @default.
- W4297688052 hasRelatedWork W3074294383 @default.
- W4297688052 hasRelatedWork W3111983280 @default.
- W4297688052 hasRelatedWork W3139193008 @default.
- W4297688052 hasRelatedWork W4206669594 @default.
- W4297688052 hasRelatedWork W4295941380 @default.
- W4297688052 hasRelatedWork W4306674287 @default.
- W4297688052 hasRelatedWork W4319083788 @default.
- W4297688052 hasVolume "9" @default.
- W4297688052 isParatext "false" @default.
- W4297688052 isRetracted "false" @default.
- W4297688052 workType "article" @default.