Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963636645> ?p ?o ?g. }
- W2963636645 abstract "Previous work in hierarchical reinforcement learning has faced a dilemma: either ignore the values of different possible exit states from a subroutine, thereby risking suboptimal behavior, or represent those values explicitly thereby incurring a possibly large representation cost because exit values refer to nonlocal aspects of the world (i.e., all subsequent rewards). This paper shows that, in many cases, one can avoid both of these problems. The solution is based on recursively decomposing the exit value function in terms of Q-functions at higher levels of the hierarchy. This leads to an intuitively appealing runtime architecture in which a parent subroutine passes to its child a value function on the exit states and the child reasons about how its choices affect the exit value. We also identify structural conditions on the value function and transition distributions that allow much more concise representations of exit state distributions, leading to further state abstraction. In essence, the only variables whose exit values need be considered are those that the parent cares about and the child affects. We demonstrate the utility of our algorithms on a series of increasingly complex environments." @default.
- W2963636645 created "2019-07-30" @default.
- W2963636645 creator A5004359965 @default.
- W2963636645 creator A5007305440 @default.
- W2963636645 creator A5021300739 @default.
- W2963636645 date "2006-01-01" @default.
- W2963636645 modified "2023-10-17" @default.
- W2963636645 title "A compact, hierarchical Q-function decomposition" @default.
- W2963636645 hasPublicationYear "2006" @default.
- W2963636645 type Work @default.
- W2963636645 sameAs 2963636645 @default.
- W2963636645 citedByCount "3" @default.
- W2963636645 countsByYear W29636366452018 @default.
- W2963636645 crossrefType "proceedings-article" @default.
- W2963636645 hasAuthorship W2963636645A5004359965 @default.
- W2963636645 hasAuthorship W2963636645A5007305440 @default.
- W2963636645 hasAuthorship W2963636645A5021300739 @default.
- W2963636645 hasConcept C103784038 @default.
- W2963636645 hasConcept C105795698 @default.
- W2963636645 hasConcept C111472728 @default.
- W2963636645 hasConcept C11413529 @default.
- W2963636645 hasConcept C119857082 @default.
- W2963636645 hasConcept C124304363 @default.
- W2963636645 hasConcept C124681953 @default.
- W2963636645 hasConcept C126255220 @default.
- W2963636645 hasConcept C138885662 @default.
- W2963636645 hasConcept C14036430 @default.
- W2963636645 hasConcept C143724316 @default.
- W2963636645 hasConcept C14646407 @default.
- W2963636645 hasConcept C151730666 @default.
- W2963636645 hasConcept C154945302 @default.
- W2963636645 hasConcept C162324750 @default.
- W2963636645 hasConcept C17744445 @default.
- W2963636645 hasConcept C18903297 @default.
- W2963636645 hasConcept C197055811 @default.
- W2963636645 hasConcept C199360897 @default.
- W2963636645 hasConcept C199539241 @default.
- W2963636645 hasConcept C2524010 @default.
- W2963636645 hasConcept C2776291640 @default.
- W2963636645 hasConcept C2776359362 @default.
- W2963636645 hasConcept C2778496695 @default.
- W2963636645 hasConcept C31170391 @default.
- W2963636645 hasConcept C33923547 @default.
- W2963636645 hasConcept C34447519 @default.
- W2963636645 hasConcept C41008148 @default.
- W2963636645 hasConcept C48103436 @default.
- W2963636645 hasConcept C55842286 @default.
- W2963636645 hasConcept C78458016 @default.
- W2963636645 hasConcept C80444323 @default.
- W2963636645 hasConcept C86803240 @default.
- W2963636645 hasConcept C94625758 @default.
- W2963636645 hasConcept C96147967 @default.
- W2963636645 hasConcept C97541855 @default.
- W2963636645 hasConceptScore W2963636645C103784038 @default.
- W2963636645 hasConceptScore W2963636645C105795698 @default.
- W2963636645 hasConceptScore W2963636645C111472728 @default.
- W2963636645 hasConceptScore W2963636645C11413529 @default.
- W2963636645 hasConceptScore W2963636645C119857082 @default.
- W2963636645 hasConceptScore W2963636645C124304363 @default.
- W2963636645 hasConceptScore W2963636645C124681953 @default.
- W2963636645 hasConceptScore W2963636645C126255220 @default.
- W2963636645 hasConceptScore W2963636645C138885662 @default.
- W2963636645 hasConceptScore W2963636645C14036430 @default.
- W2963636645 hasConceptScore W2963636645C143724316 @default.
- W2963636645 hasConceptScore W2963636645C14646407 @default.
- W2963636645 hasConceptScore W2963636645C151730666 @default.
- W2963636645 hasConceptScore W2963636645C154945302 @default.
- W2963636645 hasConceptScore W2963636645C162324750 @default.
- W2963636645 hasConceptScore W2963636645C17744445 @default.
- W2963636645 hasConceptScore W2963636645C18903297 @default.
- W2963636645 hasConceptScore W2963636645C197055811 @default.
- W2963636645 hasConceptScore W2963636645C199360897 @default.
- W2963636645 hasConceptScore W2963636645C199539241 @default.
- W2963636645 hasConceptScore W2963636645C2524010 @default.
- W2963636645 hasConceptScore W2963636645C2776291640 @default.
- W2963636645 hasConceptScore W2963636645C2776359362 @default.
- W2963636645 hasConceptScore W2963636645C2778496695 @default.
- W2963636645 hasConceptScore W2963636645C31170391 @default.
- W2963636645 hasConceptScore W2963636645C33923547 @default.
- W2963636645 hasConceptScore W2963636645C34447519 @default.
- W2963636645 hasConceptScore W2963636645C41008148 @default.
- W2963636645 hasConceptScore W2963636645C48103436 @default.
- W2963636645 hasConceptScore W2963636645C55842286 @default.
- W2963636645 hasConceptScore W2963636645C78458016 @default.
- W2963636645 hasConceptScore W2963636645C80444323 @default.
- W2963636645 hasConceptScore W2963636645C86803240 @default.
- W2963636645 hasConceptScore W2963636645C94625758 @default.
- W2963636645 hasConceptScore W2963636645C96147967 @default.
- W2963636645 hasConceptScore W2963636645C97541855 @default.
- W2963636645 hasLocation W29636366451 @default.
- W2963636645 hasOpenAccess W2963636645 @default.
- W2963636645 hasPrimaryLocation W29636366451 @default.
- W2963636645 hasRelatedWork W142858861 @default.
- W2963636645 hasRelatedWork W1505837856 @default.
- W2963636645 hasRelatedWork W1506796530 @default.
- W2963636645 hasRelatedWork W1554774506 @default.
- W2963636645 hasRelatedWork W1812382750 @default.
- W2963636645 hasRelatedWork W2031063962 @default.
- W2963636645 hasRelatedWork W2035509265 @default.
- W2963636645 hasRelatedWork W2121517924 @default.