Matches in SemOpenAlex for { <https://semopenalex.org/work/W2893250549> ?p ?o ?g. }
- W2893250549 abstract "Building deep reinforcement learning agents that can generalize and adapt to unseen environments remains a fundamental challenge for AI. This paper describes progresses on this challenge in the context of man-made environments, which are visually diverse but contain intrinsic semantic regularities. We propose a hybrid model-based and model-free approach, LEArning and Planning with Semantics (LEAPS), consisting of a multi-target sub-policy that acts on visual inputs, and a Bayesian model over semantic structures. When placed in an unseen environment, the agent plans with the semantic model to make high-level decisions, proposes the next sub-target for the sub-policy to execute, and updates the semantic model based on new observations. We perform experiments in visual navigation tasks using House3D, a 3D environment that contains diverse human-designed indoor scenes with real-world objects. LEAPS outperforms strong baselines that do not explicitly plan using the semantic content." @default.
- W2893250549 created "2018-10-05" @default.
- W2893250549 creator A5007305440 @default.
- W2893250549 creator A5008951080 @default.
- W2893250549 creator A5014407395 @default.
- W2893250549 creator A5018369948 @default.
- W2893250549 creator A5020699288 @default.
- W2893250549 creator A5084821923 @default.
- W2893250549 date "2018-09-27" @default.
- W2893250549 modified "2023-09-27" @default.
- W2893250549 title "Learning and Planning with a Semantic Model" @default.
- W2893250549 cites W195596278 @default.
- W2893250549 cites W2118688707 @default.
- W2893250549 cites W2145339207 @default.
- W2893250549 cites W2257979135 @default.
- W2893250549 cites W2258731934 @default.
- W2893250549 cites W2336416123 @default.
- W2893250549 cites W2523728418 @default.
- W2893250549 cites W2528489519 @default.
- W2893250549 cites W2557465155 @default.
- W2893250549 cites W2578206533 @default.
- W2893250549 cites W2594903727 @default.
- W2893250549 cites W2604763608 @default.
- W2893250549 cites W2624780181 @default.
- W2893250549 cites W2754261200 @default.
- W2893250549 cites W2765349170 @default.
- W2893250549 cites W2783375473 @default.
- W2893250549 cites W2786122328 @default.
- W2893250549 cites W2788781499 @default.
- W2893250549 cites W2806422867 @default.
- W2893250549 cites W2949267040 @default.
- W2893250549 cites W2950872548 @default.
- W2893250549 cites W2951660448 @default.
- W2893250549 cites W2951881474 @default.
- W2893250549 cites W2952606116 @default.
- W2893250549 cites W2952791429 @default.
- W2893250549 cites W2953326374 @default.
- W2893250549 cites W2962887844 @default.
- W2893250549 cites W2963305465 @default.
- W2893250549 cites W2963946945 @default.
- W2893250549 cites W2964043796 @default.
- W2893250549 cites W2964161785 @default.
- W2893250549 hasPublicationYear "2018" @default.
- W2893250549 type Work @default.
- W2893250549 sameAs 2893250549 @default.
- W2893250549 citedByCount "8" @default.
- W2893250549 countsByYear W28932505492019 @default.
- W2893250549 countsByYear W28932505492020 @default.
- W2893250549 countsByYear W28932505492021 @default.
- W2893250549 crossrefType "posted-content" @default.
- W2893250549 hasAuthorship W2893250549A5007305440 @default.
- W2893250549 hasAuthorship W2893250549A5008951080 @default.
- W2893250549 hasAuthorship W2893250549A5014407395 @default.
- W2893250549 hasAuthorship W2893250549A5018369948 @default.
- W2893250549 hasAuthorship W2893250549A5020699288 @default.
- W2893250549 hasAuthorship W2893250549A5084821923 @default.
- W2893250549 hasConcept C106159729 @default.
- W2893250549 hasConcept C107457646 @default.
- W2893250549 hasConcept C119857082 @default.
- W2893250549 hasConcept C151730666 @default.
- W2893250549 hasConcept C154945302 @default.
- W2893250549 hasConcept C162324750 @default.
- W2893250549 hasConcept C166957645 @default.
- W2893250549 hasConcept C184337299 @default.
- W2893250549 hasConcept C199360897 @default.
- W2893250549 hasConcept C2776505523 @default.
- W2893250549 hasConcept C2777705401 @default.
- W2893250549 hasConcept C2779343474 @default.
- W2893250549 hasConcept C41008148 @default.
- W2893250549 hasConcept C86803240 @default.
- W2893250549 hasConcept C90312973 @default.
- W2893250549 hasConcept C95457728 @default.
- W2893250549 hasConcept C97541855 @default.
- W2893250549 hasConceptScore W2893250549C106159729 @default.
- W2893250549 hasConceptScore W2893250549C107457646 @default.
- W2893250549 hasConceptScore W2893250549C119857082 @default.
- W2893250549 hasConceptScore W2893250549C151730666 @default.
- W2893250549 hasConceptScore W2893250549C154945302 @default.
- W2893250549 hasConceptScore W2893250549C162324750 @default.
- W2893250549 hasConceptScore W2893250549C166957645 @default.
- W2893250549 hasConceptScore W2893250549C184337299 @default.
- W2893250549 hasConceptScore W2893250549C199360897 @default.
- W2893250549 hasConceptScore W2893250549C2776505523 @default.
- W2893250549 hasConceptScore W2893250549C2777705401 @default.
- W2893250549 hasConceptScore W2893250549C2779343474 @default.
- W2893250549 hasConceptScore W2893250549C41008148 @default.
- W2893250549 hasConceptScore W2893250549C86803240 @default.
- W2893250549 hasConceptScore W2893250549C90312973 @default.
- W2893250549 hasConceptScore W2893250549C95457728 @default.
- W2893250549 hasConceptScore W2893250549C97541855 @default.
- W2893250549 hasLocation W28932505491 @default.
- W2893250549 hasOpenAccess W2893250549 @default.
- W2893250549 hasPrimaryLocation W28932505491 @default.
- W2893250549 hasRelatedWork W2194775991 @default.
- W2893250549 hasRelatedWork W2567015638 @default.
- W2893250549 hasRelatedWork W2593841437 @default.
- W2893250549 hasRelatedWork W2776202271 @default.
- W2893250549 hasRelatedWork W2884565639 @default.
- W2893250549 hasRelatedWork W2951550102 @default.
- W2893250549 hasRelatedWork W2962732398 @default.