Matches in SemOpenAlex for { <https://semopenalex.org/work/W2971234462> ?p ?o ?g. }
- W2971234462 endingPage "14144" @default.
- W2971234462 startingPage "14134" @default.
- W2971234462 abstract "In an effort to better understand the different ways in which the discount factor affects the optimization process in reinforcement learning, we designed a set of experiments to study each effect in isolation. Our analysis reveals that the common perception that poor performance of low discount factors is caused by (too) small action-gaps requires revision. We propose an alternative hypothesis that identifies the size-difference of the action-gap across the state-space as the primary cause. We then introduce a new method that enables more homogeneous action-gaps by mapping value estimates to a logarithmic space. We prove convergence for this method under standard assumptions and demonstrate empirically that it indeed enables lower discount factors for approximate reinforcement-learning methods. This in turn allows tackling a class of reinforcement-learning problems that are challenging to solve with traditional methods." @default.
- W2971234462 created "2019-09-05" @default.
- W2971234462 creator A5049195732 @default.
- W2971234462 creator A5070690406 @default.
- W2971234462 creator A5081617594 @default.
- W2971234462 date "2019-12-01" @default.
- W2971234462 modified "2023-09-24" @default.
- W2971234462 title "Using a Logarithmic Mapping to Enable Lower Discount Factors in Reinforcement Learning" @default.
- W2971234462 hasPublicationYear "2019" @default.
- W2971234462 type Work @default.
- W2971234462 sameAs 2971234462 @default.
- W2971234462 citedByCount "11" @default.
- W2971234462 countsByYear W29712344622020 @default.
- W2971234462 countsByYear W29712344622021 @default.
- W2971234462 crossrefType "proceedings-article" @default.
- W2971234462 hasAuthorship W2971234462A5049195732 @default.
- W2971234462 hasAuthorship W2971234462A5070690406 @default.
- W2971234462 hasAuthorship W2971234462A5081617594 @default.
- W2971234462 hasConcept C10138342 @default.
- W2971234462 hasConcept C105795698 @default.
- W2971234462 hasConcept C106189395 @default.
- W2971234462 hasConcept C111919701 @default.
- W2971234462 hasConcept C119857082 @default.
- W2971234462 hasConcept C121332964 @default.
- W2971234462 hasConcept C127413603 @default.
- W2971234462 hasConcept C134306372 @default.
- W2971234462 hasConcept C154945302 @default.
- W2971234462 hasConcept C159886148 @default.
- W2971234462 hasConcept C162324750 @default.
- W2971234462 hasConcept C177264268 @default.
- W2971234462 hasConcept C196340769 @default.
- W2971234462 hasConcept C199360897 @default.
- W2971234462 hasConcept C2777212361 @default.
- W2971234462 hasConcept C2777303404 @default.
- W2971234462 hasConcept C2778572836 @default.
- W2971234462 hasConcept C2780791683 @default.
- W2971234462 hasConcept C33923547 @default.
- W2971234462 hasConcept C39927690 @default.
- W2971234462 hasConcept C41008148 @default.
- W2971234462 hasConcept C50522688 @default.
- W2971234462 hasConcept C6177178 @default.
- W2971234462 hasConcept C62520636 @default.
- W2971234462 hasConcept C66938386 @default.
- W2971234462 hasConcept C67203356 @default.
- W2971234462 hasConcept C72434380 @default.
- W2971234462 hasConcept C97541855 @default.
- W2971234462 hasConceptScore W2971234462C10138342 @default.
- W2971234462 hasConceptScore W2971234462C105795698 @default.
- W2971234462 hasConceptScore W2971234462C106189395 @default.
- W2971234462 hasConceptScore W2971234462C111919701 @default.
- W2971234462 hasConceptScore W2971234462C119857082 @default.
- W2971234462 hasConceptScore W2971234462C121332964 @default.
- W2971234462 hasConceptScore W2971234462C127413603 @default.
- W2971234462 hasConceptScore W2971234462C134306372 @default.
- W2971234462 hasConceptScore W2971234462C154945302 @default.
- W2971234462 hasConceptScore W2971234462C159886148 @default.
- W2971234462 hasConceptScore W2971234462C162324750 @default.
- W2971234462 hasConceptScore W2971234462C177264268 @default.
- W2971234462 hasConceptScore W2971234462C196340769 @default.
- W2971234462 hasConceptScore W2971234462C199360897 @default.
- W2971234462 hasConceptScore W2971234462C2777212361 @default.
- W2971234462 hasConceptScore W2971234462C2777303404 @default.
- W2971234462 hasConceptScore W2971234462C2778572836 @default.
- W2971234462 hasConceptScore W2971234462C2780791683 @default.
- W2971234462 hasConceptScore W2971234462C33923547 @default.
- W2971234462 hasConceptScore W2971234462C39927690 @default.
- W2971234462 hasConceptScore W2971234462C41008148 @default.
- W2971234462 hasConceptScore W2971234462C50522688 @default.
- W2971234462 hasConceptScore W2971234462C6177178 @default.
- W2971234462 hasConceptScore W2971234462C62520636 @default.
- W2971234462 hasConceptScore W2971234462C66938386 @default.
- W2971234462 hasConceptScore W2971234462C67203356 @default.
- W2971234462 hasConceptScore W2971234462C72434380 @default.
- W2971234462 hasConceptScore W2971234462C97541855 @default.
- W2971234462 hasLocation W29712344621 @default.
- W2971234462 hasOpenAccess W2971234462 @default.
- W2971234462 hasPrimaryLocation W29712344621 @default.
- W2971234462 hasRelatedWork W142858861 @default.
- W2971234462 hasRelatedWork W1616330627 @default.
- W2971234462 hasRelatedWork W1771410628 @default.
- W2971234462 hasRelatedWork W2121863487 @default.
- W2971234462 hasRelatedWork W2145339207 @default.
- W2971234462 hasRelatedWork W2155027007 @default.
- W2971234462 hasRelatedWork W2158782408 @default.
- W2971234462 hasRelatedWork W2224508600 @default.
- W2971234462 hasRelatedWork W2575066953 @default.
- W2971234462 hasRelatedWork W2736601468 @default.
- W2971234462 hasRelatedWork W2773896370 @default.
- W2971234462 hasRelatedWork W2787477569 @default.
- W2971234462 hasRelatedWork W2890211540 @default.
- W2971234462 hasRelatedWork W2902854091 @default.
- W2971234462 hasRelatedWork W2947241716 @default.
- W2971234462 hasRelatedWork W2964043796 @default.
- W2971234462 hasRelatedWork W2970202659 @default.
- W2971234462 hasRelatedWork W2979895842 @default.
- W2971234462 hasRelatedWork W3205279445 @default.
- W2971234462 hasRelatedWork W3124153248 @default.
- W2971234462 hasVolume "32" @default.