Matches in SemOpenAlex for { <https://semopenalex.org/work/W3209719579> ?p ?o ?g. }
- W3209719579 abstract "Principled decision-making in continuous state--action spaces is impossible without some assumptions. A common approach is to assume Lipschitz continuity of the Q-function. We show that, unfortunately, this property fails to hold in many typical domains. We propose a new coarse-grained smoothness definition that generalizes the notion of Lipschitz continuity, is more widely applicable, and allows us to compute significantly tighter bounds on Q-functions, leading to improved learning. We provide a theoretical analysis of our new smoothness definition, and discuss its implications and impact on control and exploration in continuous domains." @default.
- W3209719579 created "2021-11-08" @default.
- W3209719579 creator A5009722403 @default.
- W3209719579 creator A5016921301 @default.
- W3209719579 creator A5037667167 @default.
- W3209719579 creator A5045518550 @default.
- W3209719579 creator A5061301686 @default.
- W3209719579 creator A5078124517 @default.
- W3209719579 date "2021-10-23" @default.
- W3209719579 modified "2023-09-23" @default.
- W3209719579 title "Coarse-Grained Smoothness for RL in Metric Spaces." @default.
- W3209719579 cites W1701974503 @default.
- W3209719579 cites W1747856733 @default.
- W3209719579 cites W1988526405 @default.
- W3209719579 cites W2046859786 @default.
- W3209719579 cites W2103708221 @default.
- W3209719579 cites W2121863487 @default.
- W3209719579 cites W21891419 @default.
- W3209719579 cites W2489939061 @default.
- W3209719579 cites W2739748921 @default.
- W3209719579 cites W2894842463 @default.
- W3209719579 cites W2943357710 @default.
- W3209719579 cites W2943632322 @default.
- W3209719579 cites W2946529464 @default.
- W3209719579 cites W2950320358 @default.
- W3209719579 cites W2951333509 @default.
- W3209719579 cites W2952905979 @default.
- W3209719579 cites W2962857907 @default.
- W3209719579 cites W2963395712 @default.
- W3209719579 cites W2963488874 @default.
- W3209719579 cites W2965300894 @default.
- W3209719579 cites W2992545808 @default.
- W3209719579 cites W2999710481 @default.
- W3209719579 cites W3009498344 @default.
- W3209719579 cites W3038887435 @default.
- W3209719579 cites W3094117125 @default.
- W3209719579 hasPublicationYear "2021" @default.
- W3209719579 type Work @default.
- W3209719579 sameAs 3209719579 @default.
- W3209719579 citedByCount "0" @default.
- W3209719579 crossrefType "posted-content" @default.
- W3209719579 hasAuthorship W3209719579A5009722403 @default.
- W3209719579 hasAuthorship W3209719579A5016921301 @default.
- W3209719579 hasAuthorship W3209719579A5037667167 @default.
- W3209719579 hasAuthorship W3209719579A5045518550 @default.
- W3209719579 hasAuthorship W3209719579A5061301686 @default.
- W3209719579 hasAuthorship W3209719579A5078124517 @default.
- W3209719579 hasConcept C102634674 @default.
- W3209719579 hasConcept C111472728 @default.
- W3209719579 hasConcept C11413529 @default.
- W3209719579 hasConcept C121332964 @default.
- W3209719579 hasConcept C126255220 @default.
- W3209719579 hasConcept C134306372 @default.
- W3209719579 hasConcept C138885662 @default.
- W3209719579 hasConcept C14036430 @default.
- W3209719579 hasConcept C144237770 @default.
- W3209719579 hasConcept C162324750 @default.
- W3209719579 hasConcept C176217482 @default.
- W3209719579 hasConcept C189950617 @default.
- W3209719579 hasConcept C198043062 @default.
- W3209719579 hasConcept C202444582 @default.
- W3209719579 hasConcept C21547014 @default.
- W3209719579 hasConcept C22324862 @default.
- W3209719579 hasConcept C2780791683 @default.
- W3209719579 hasConcept C28826006 @default.
- W3209719579 hasConcept C33923547 @default.
- W3209719579 hasConcept C41008148 @default.
- W3209719579 hasConcept C48103436 @default.
- W3209719579 hasConcept C62520636 @default.
- W3209719579 hasConcept C78458016 @default.
- W3209719579 hasConcept C86803240 @default.
- W3209719579 hasConceptScore W3209719579C102634674 @default.
- W3209719579 hasConceptScore W3209719579C111472728 @default.
- W3209719579 hasConceptScore W3209719579C11413529 @default.
- W3209719579 hasConceptScore W3209719579C121332964 @default.
- W3209719579 hasConceptScore W3209719579C126255220 @default.
- W3209719579 hasConceptScore W3209719579C134306372 @default.
- W3209719579 hasConceptScore W3209719579C138885662 @default.
- W3209719579 hasConceptScore W3209719579C14036430 @default.
- W3209719579 hasConceptScore W3209719579C144237770 @default.
- W3209719579 hasConceptScore W3209719579C162324750 @default.
- W3209719579 hasConceptScore W3209719579C176217482 @default.
- W3209719579 hasConceptScore W3209719579C189950617 @default.
- W3209719579 hasConceptScore W3209719579C198043062 @default.
- W3209719579 hasConceptScore W3209719579C202444582 @default.
- W3209719579 hasConceptScore W3209719579C21547014 @default.
- W3209719579 hasConceptScore W3209719579C22324862 @default.
- W3209719579 hasConceptScore W3209719579C2780791683 @default.
- W3209719579 hasConceptScore W3209719579C28826006 @default.
- W3209719579 hasConceptScore W3209719579C33923547 @default.
- W3209719579 hasConceptScore W3209719579C41008148 @default.
- W3209719579 hasConceptScore W3209719579C48103436 @default.
- W3209719579 hasConceptScore W3209719579C62520636 @default.
- W3209719579 hasConceptScore W3209719579C78458016 @default.
- W3209719579 hasConceptScore W3209719579C86803240 @default.
- W3209719579 hasOpenAccess W3209719579 @default.
- W3209719579 hasRelatedWork W1557570059 @default.
- W3209719579 hasRelatedWork W1883031669 @default.
- W3209719579 hasRelatedWork W2011857371 @default.
- W3209719579 hasRelatedWork W2041281474 @default.