Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963557365> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W2963557365 endingPage "3119" @default.
- W2963557365 startingPage "3110" @default.
- W2963557365 abstract "Deep reinforcement learning (DRL) has achieved significant breakthroughs in various tasks. However, most DRL algorithms suffer a problem of generalising the learned policy, which makes the policy performance largely affected even by minor modifications of the training environment. Except that, the use of deep neural networks makes the learned policies hard to be interpretable. To address these two challenges, we propose a novel algorithm named Neural Logic Reinforcement Learning (NLRL) to represent the policies in reinforcement learning by first-order logic. NLRL is based on policy gradient methods and differentiable inductive logic programming that have demonstrated significant advantages in terms of interpretability and generalisability in supervised tasks. Extensive experiments conducted on cliff-walking and blocks manipulation tasks demonstrate that NLRL can induce interpretable policies achieving near-optimal performance while showing good generalisability to environments of different initial states and problem sizes." @default.
- W2963557365 created "2019-07-30" @default.
- W2963557365 creator A5012646628 @default.
- W2963557365 creator A5085148839 @default.
- W2963557365 date "2019-06-10" @default.
- W2963557365 modified "2023-09-27" @default.
- W2963557365 title "Neural Logic Reinforcement Learning" @default.
- W2963557365 hasPublicationYear "2019" @default.
- W2963557365 type Work @default.
- W2963557365 sameAs 2963557365 @default.
- W2963557365 citedByCount "13" @default.
- W2963557365 countsByYear W29635573652019 @default.
- W2963557365 countsByYear W29635573652020 @default.
- W2963557365 countsByYear W29635573652021 @default.
- W2963557365 countsByYear W29635573652022 @default.
- W2963557365 crossrefType "proceedings-article" @default.
- W2963557365 hasAuthorship W2963557365A5012646628 @default.
- W2963557365 hasAuthorship W2963557365A5085148839 @default.
- W2963557365 hasConcept C119857082 @default.
- W2963557365 hasConcept C134306372 @default.
- W2963557365 hasConcept C154945302 @default.
- W2963557365 hasConcept C202615002 @default.
- W2963557365 hasConcept C2779382394 @default.
- W2963557365 hasConcept C2781067378 @default.
- W2963557365 hasConcept C2984842247 @default.
- W2963557365 hasConcept C33923547 @default.
- W2963557365 hasConcept C41008148 @default.
- W2963557365 hasConcept C50644808 @default.
- W2963557365 hasConcept C97541855 @default.
- W2963557365 hasConceptScore W2963557365C119857082 @default.
- W2963557365 hasConceptScore W2963557365C134306372 @default.
- W2963557365 hasConceptScore W2963557365C154945302 @default.
- W2963557365 hasConceptScore W2963557365C202615002 @default.
- W2963557365 hasConceptScore W2963557365C2779382394 @default.
- W2963557365 hasConceptScore W2963557365C2781067378 @default.
- W2963557365 hasConceptScore W2963557365C2984842247 @default.
- W2963557365 hasConceptScore W2963557365C33923547 @default.
- W2963557365 hasConceptScore W2963557365C41008148 @default.
- W2963557365 hasConceptScore W2963557365C50644808 @default.
- W2963557365 hasConceptScore W2963557365C97541855 @default.
- W2963557365 hasLocation W29635573651 @default.
- W2963557365 hasOpenAccess W2963557365 @default.
- W2963557365 hasPrimaryLocation W29635573651 @default.
- W2963557365 hasRelatedWork W1595483645 @default.
- W2963557365 hasRelatedWork W2145339207 @default.
- W2963557365 hasRelatedWork W2521274174 @default.
- W2963557365 hasRelatedWork W2735995851 @default.
- W2963557365 hasRelatedWork W276460289 @default.
- W2963557365 hasRelatedWork W2892858663 @default.
- W2963557365 hasRelatedWork W2940734525 @default.
- W2963557365 hasRelatedWork W2946824041 @default.
- W2963557365 hasRelatedWork W2962924847 @default.
- W2963557365 hasRelatedWork W2963286043 @default.
- W2963557365 hasRelatedWork W2964464273 @default.
- W2963557365 hasRelatedWork W2971919016 @default.
- W2963557365 hasRelatedWork W3010862467 @default.
- W2963557365 hasRelatedWork W3020831056 @default.
- W2963557365 hasRelatedWork W3035521307 @default.
- W2963557365 hasRelatedWork W3109409708 @default.
- W2963557365 hasRelatedWork W3131310681 @default.
- W2963557365 hasRelatedWork W3196302130 @default.
- W2963557365 hasRelatedWork W3201666735 @default.
- W2963557365 hasRelatedWork W3208540476 @default.
- W2963557365 isParatext "false" @default.
- W2963557365 isRetracted "false" @default.
- W2963557365 magId "2963557365" @default.
- W2963557365 workType "article" @default.