Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963244606> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2963244606 abstract "In this work, we study the credit assignment problem in reward augmented maximum likelihood (RAML) learning, and establish a theoretical equivalence between the token-level counterpart of RAML and the entropy regularized reinforcement learning. Inspired by the connection, we propose two sequence prediction algorithms, one extending RAML with fine-grained credit assignment and the other improving Actor-Critic with a systematic entropy regularization. On two benchmark datasets, we show the proposed algorithms outperform RAML and Actor-Critic respectively, providing new alternatives to sequence prediction." @default.
- W2963244606 created "2019-07-30" @default.
- W2963244606 creator A5029288401 @default.
- W2963244606 creator A5060225743 @default.
- W2963244606 creator A5091869105 @default.
- W2963244606 date "2018-01-01" @default.
- W2963244606 modified "2023-09-26" @default.
- W2963244606 title "From Credit Assignment to Entropy Regularization: Two New Algorithms for Neural Sequence Prediction" @default.
- W2963244606 cites W1514535095 @default.
- W2963244606 cites W1646707810 @default.
- W2963244606 cites W1861492603 @default.
- W2963244606 cites W1895577753 @default.
- W2963244606 cites W1902237438 @default.
- W2963244606 cites W1993411524 @default.
- W2963244606 cites W2064675550 @default.
- W2963244606 cites W2098774185 @default.
- W2963244606 cites W2101105183 @default.
- W2963244606 cites W2130942839 @default.
- W2963244606 cites W2145339207 @default.
- W2963244606 cites W2173248099 @default.
- W2963244606 cites W2176263492 @default.
- W2963244606 cites W2184135559 @default.
- W2963244606 cites W2194775991 @default.
- W2963244606 cites W2327501763 @default.
- W2963244606 cites W2487501366 @default.
- W2963244606 cites W2508728158 @default.
- W2963244606 cites W2542835211 @default.
- W2963244606 cites W2580192806 @default.
- W2963244606 cites W2593044849 @default.
- W2963244606 cites W2594103415 @default.
- W2963244606 cites W2609650878 @default.
- W2963244606 cites W2618606525 @default.
- W2963244606 cites W2781726626 @default.
- W2963244606 cites W2962965405 @default.
- W2963244606 cites W2963084599 @default.
- W2963244606 cites W2963463964 @default.
- W2963244606 cites W2963620441 @default.
- W2963244606 cites W2964043796 @default.
- W2963244606 cites W2964308564 @default.
- W2963244606 cites W64088143 @default.
- W2963244606 cites W648786980 @default.
- W2963244606 cites W2010624529 @default.
- W2963244606 doi "https://doi.org/10.18653/v1/p18-1155" @default.
- W2963244606 hasPublicationYear "2018" @default.
- W2963244606 type Work @default.
- W2963244606 sameAs 2963244606 @default.
- W2963244606 citedByCount "1" @default.
- W2963244606 countsByYear W29632446062020 @default.
- W2963244606 crossrefType "proceedings-article" @default.
- W2963244606 hasAuthorship W2963244606A5029288401 @default.
- W2963244606 hasAuthorship W2963244606A5060225743 @default.
- W2963244606 hasAuthorship W2963244606A5091869105 @default.
- W2963244606 hasBestOaLocation W29632446061 @default.
- W2963244606 hasConcept C106301342 @default.
- W2963244606 hasConcept C11413529 @default.
- W2963244606 hasConcept C121332964 @default.
- W2963244606 hasConcept C154945302 @default.
- W2963244606 hasConcept C185592680 @default.
- W2963244606 hasConcept C2776135515 @default.
- W2963244606 hasConcept C2778112365 @default.
- W2963244606 hasConcept C41008148 @default.
- W2963244606 hasConcept C50644808 @default.
- W2963244606 hasConcept C55493867 @default.
- W2963244606 hasConcept C97355855 @default.
- W2963244606 hasConceptScore W2963244606C106301342 @default.
- W2963244606 hasConceptScore W2963244606C11413529 @default.
- W2963244606 hasConceptScore W2963244606C121332964 @default.
- W2963244606 hasConceptScore W2963244606C154945302 @default.
- W2963244606 hasConceptScore W2963244606C185592680 @default.
- W2963244606 hasConceptScore W2963244606C2776135515 @default.
- W2963244606 hasConceptScore W2963244606C2778112365 @default.
- W2963244606 hasConceptScore W2963244606C41008148 @default.
- W2963244606 hasConceptScore W2963244606C50644808 @default.
- W2963244606 hasConceptScore W2963244606C55493867 @default.
- W2963244606 hasConceptScore W2963244606C97355855 @default.
- W2963244606 hasLocation W29632446061 @default.
- W2963244606 hasLocation W29632446062 @default.
- W2963244606 hasOpenAccess W2963244606 @default.
- W2963244606 hasPrimaryLocation W29632446061 @default.
- W2963244606 hasRelatedWork W2057419801 @default.
- W2963244606 hasRelatedWork W2351491280 @default.
- W2963244606 hasRelatedWork W2371447506 @default.
- W2963244606 hasRelatedWork W2380313759 @default.
- W2963244606 hasRelatedWork W2386387936 @default.
- W2963244606 hasRelatedWork W2386767533 @default.
- W2963244606 hasRelatedWork W2392110728 @default.
- W2963244606 hasRelatedWork W303980170 @default.
- W2963244606 hasRelatedWork W3107474891 @default.
- W2963244606 hasRelatedWork W1629725936 @default.
- W2963244606 isParatext "false" @default.
- W2963244606 isRetracted "false" @default.
- W2963244606 magId "2963244606" @default.
- W2963244606 workType "article" @default.