FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2987753679> ?p ?o ?g. }

• W2987753679 abstract "Imitation learning trains a policy from expert demonstrations. Imitation learning approaches have been designed from various principles, such as behavioral cloning via supervised learning, apprenticeship learning via inverse reinforcement learning, and GAIL via generative adversarial learning. In this paper, we propose a framework to analyze the theoretical property of imitation learning approaches based on discrepancy propagation analysis. Under the infinite-horizon setting, the framework leads to the value discrepancy of behavioral cloning in an order of O((1-gamma)^{-2}). We also show that the framework leads to the value discrepancy of GAIL in an order of O((1-gamma)^{-1}). It implies that GAIL has less compounding errors than behavioral cloning, which is also verified empirically in this paper. To the best of our knowledge, we are the first one to analyze GAIL's performance theoretically. The above results indicate that the proposed framework is a general tool to analyze imitation learning approaches. We hope our theoretical results can provide insights for future improvements in imitation learning algorithms." @default.
• W2987753679 created "2019-11-22" @default.
• W2987753679 creator A5016024688 @default.
• W2987753679 creator A5045734626 @default.
• W2987753679 creator A5067535440 @default.
• W2987753679 date "2019-11-16" @default.
• W2987753679 modified "2023-09-27" @default.
• W2987753679 title "On Value Discrepancy of Imitation Learning." @default.
• W2987753679 cites W1515851193 @default.
• W2987753679 cites W1975463331 @default.
• W2987753679 cites W1999874108 @default.
• W2987753679 cites W2061562262 @default.
• W2987753679 cites W2098774185 @default.
• W2987753679 cites W2099471712 @default.
• W2987753679 cites W2113023245 @default.
• W2987753679 cites W2147544021 @default.
• W2987753679 cites W2174424190 @default.
• W2987753679 cites W2290104316 @default.
• W2987753679 cites W2342840547 @default.
• W2987753679 cites W2401592218 @default.
• W2987753679 cites W2434014514 @default.
• W2987753679 cites W2489939061 @default.
• W2987753679 cites W2952745707 @default.
• W2987753679 cites W2962787969 @default.
• W2987753679 cites W2962848977 @default.
• W2987753679 cites W2962957031 @default.
• W2987753679 cites W2963508354 @default.
• W2987753679 cites W2964209830 @default.
• W2987753679 cites W2971001803 @default.
• W2987753679 hasPublicationYear "2019" @default.
• W2987753679 type Work @default.
• W2987753679 sameAs 2987753679 @default.
• W2987753679 citedByCount "2" @default.
• W2987753679 countsByYear W29877536792021 @default.
• W2987753679 crossrefType "posted-content" @default.
• W2987753679 hasAuthorship W2987753679A5016024688 @default.
• W2987753679 hasAuthorship W2987753679A5045734626 @default.
• W2987753679 hasAuthorship W2987753679A5067535440 @default.
• W2987753679 hasConcept C10138342 @default.
• W2987753679 hasConcept C111472728 @default.
• W2987753679 hasConcept C119857082 @default.
• W2987753679 hasConcept C121050878 @default.
• W2987753679 hasConcept C126388530 @default.
• W2987753679 hasConcept C138885662 @default.
• W2987753679 hasConcept C154945302 @default.
• W2987753679 hasConcept C15744967 @default.
• W2987753679 hasConcept C162324750 @default.
• W2987753679 hasConcept C182306322 @default.
• W2987753679 hasConcept C189950617 @default.
• W2987753679 hasConcept C199360897 @default.
• W2987753679 hasConcept C2776291640 @default.
• W2987753679 hasConcept C37736160 @default.
• W2987753679 hasConcept C39890363 @default.
• W2987753679 hasConcept C40506919 @default.
• W2987753679 hasConcept C41008148 @default.
• W2987753679 hasConcept C77805123 @default.
• W2987753679 hasConcept C97541855 @default.
• W2987753679 hasConceptScore W2987753679C10138342 @default.
• W2987753679 hasConceptScore W2987753679C111472728 @default.
• W2987753679 hasConceptScore W2987753679C119857082 @default.
• W2987753679 hasConceptScore W2987753679C121050878 @default.
• W2987753679 hasConceptScore W2987753679C126388530 @default.
• W2987753679 hasConceptScore W2987753679C138885662 @default.
• W2987753679 hasConceptScore W2987753679C154945302 @default.
• W2987753679 hasConceptScore W2987753679C15744967 @default.
• W2987753679 hasConceptScore W2987753679C162324750 @default.
• W2987753679 hasConceptScore W2987753679C182306322 @default.
• W2987753679 hasConceptScore W2987753679C189950617 @default.
• W2987753679 hasConceptScore W2987753679C199360897 @default.
• W2987753679 hasConceptScore W2987753679C2776291640 @default.
• W2987753679 hasConceptScore W2987753679C37736160 @default.
• W2987753679 hasConceptScore W2987753679C39890363 @default.
• W2987753679 hasConceptScore W2987753679C40506919 @default.
• W2987753679 hasConceptScore W2987753679C41008148 @default.
• W2987753679 hasConceptScore W2987753679C77805123 @default.
• W2987753679 hasConceptScore W2987753679C97541855 @default.
• W2987753679 hasLocation W29877536791 @default.
• W2987753679 hasOpenAccess W2987753679 @default.
• W2987753679 hasPrimaryLocation W29877536791 @default.
• W2987753679 hasRelatedWork W114592305 @default.
• W2987753679 hasRelatedWork W1595666167 @default.
• W2987753679 hasRelatedWork W1621791442 @default.
• W2987753679 hasRelatedWork W1985388911 @default.
• W2987753679 hasRelatedWork W1987812720 @default.
• W2987753679 hasRelatedWork W2009656295 @default.
• W2987753679 hasRelatedWork W2084291105 @default.
• W2987753679 hasRelatedWork W2101921692 @default.
• W2987753679 hasRelatedWork W2164035091 @default.
• W2987753679 hasRelatedWork W2169537414 @default.
• W2987753679 hasRelatedWork W2271262588 @default.
• W2987753679 hasRelatedWork W2945791024 @default.
• W2987753679 hasRelatedWork W295306427 @default.
• W2987753679 hasRelatedWork W2963423916 @default.
• W2987753679 hasRelatedWork W2985308740 @default.
• W2987753679 hasRelatedWork W3011112921 @default.
• W2987753679 hasRelatedWork W3089538634 @default.
• W2987753679 hasRelatedWork W3094806401 @default.
• W2987753679 hasRelatedWork W3198998390 @default.
• W2987753679 hasRelatedWork W3202826493 @default.
• W2987753679 isParatext "false" @default.

• next