FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201628275> ?p ?o ?g. }

• W3201628275 endingPage "8158" @default.
• W3201628275 startingPage "8150" @default.
• W3201628275 abstract "Deep learning has redefined AI thanks to the rise of artificial neural networks, which are inspired by neuronal networks in the brain. Through the years, these interactions between AI and neuroscience have brought immense benefits to both fields, allowing neural networks to be used in a plethora of applications. Neural networks use an efficient implementation of reverse differentiation, called backpropagation (BP). This algorithm, however, is often criticized for its biological implausibility (e.g., lack of local update rules for the parameters). Therefore, biologically plausible learning methods that rely on predictive coding (PC), a framework for describing information processing in the brain, are increasingly studied. Recent works prove that these methods can approximate BP up to a certain margin on multilayer perceptrons (MLPs), and asymptotically on any other complex model, and that zerodivergence inference learning (Z-IL), a variant of PC, is able to exactly implement BP on MLPs. However, the recent literature shows also that there is no biologically plausible method yet that can exactly replicate the weight update of BP on complex models. To fill this gap, in this paper, we generalize (PC and) Z-IL by directly defining it on computational graphs, and show that it can perform exact reverse differentiation. What results is the first PC (and so biologically plausible) algorithm that is equivalent to BP in the way of updating parameters on any neural network, providing a bridge between the interdisciplinary research of neuroscience and deep learning. Furthermore, the above results in particular also immediately provide a novel local and parallel implementation of BP." @default.
• W3201628275 created "2021-09-27" @default.
• W3201628275 creator A5004775934 @default.
• W3201628275 creator A5022889808 @default.
• W3201628275 creator A5031081461 @default.
• W3201628275 creator A5049095056 @default.
• W3201628275 creator A5091549352 @default.
• W3201628275 date "2022-06-28" @default.
• W3201628275 modified "2023-10-05" @default.
• W3201628275 title "Reverse Differentiation via Predictive Coding" @default.
• W3201628275 cites W1973845142 @default.
• W3201628275 cites W2003357516 @default.
• W3201628275 cites W2064675550 @default.
• W3201628275 cites W2119885245 @default.
• W3201628275 cites W2120079537 @default.
• W3201628275 cites W2127958135 @default.
• W3201628275 cites W2131352719 @default.
• W3201628275 cites W2134664762 @default.
• W3201628275 cites W2176287621 @default.
• W3201628275 cites W2237059274 @default.
• W3201628275 cites W2257979135 @default.
• W3201628275 cites W2482290902 @default.
• W3201628275 cites W2527798464 @default.
• W3201628275 cites W2547057105 @default.
• W3201628275 cites W2552737632 @default.
• W3201628275 cites W2604086949 @default.
• W3201628275 cites W2766447205 @default.
• W3201628275 cites W2905333383 @default.
• W3201628275 cites W2913907987 @default.
• W3201628275 cites W2944546415 @default.
• W3201628275 cites W2964017885 @default.
• W3201628275 cites W2978368159 @default.
• W3201628275 cites W3005798058 @default.
• W3201628275 cites W3006966652 @default.
• W3201628275 cites W3016391357 @default.
• W3201628275 cites W3033108376 @default.
• W3201628275 cites W3100445868 @default.
• W3201628275 doi "https://doi.org/10.1609/aaai.v36i7.20788" @default.
• W3201628275 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37205168" @default.
• W3201628275 hasPublicationYear "2022" @default.
• W3201628275 type Work @default.
• W3201628275 sameAs 3201628275 @default.
• W3201628275 citedByCount "4" @default.
• W3201628275 countsByYear W32016282752022 @default.
• W3201628275 countsByYear W32016282752023 @default.
• W3201628275 crossrefType "journal-article" @default.
• W3201628275 hasAuthorship W3201628275A5004775934 @default.
• W3201628275 hasAuthorship W3201628275A5022889808 @default.
• W3201628275 hasAuthorship W3201628275A5031081461 @default.
• W3201628275 hasAuthorship W3201628275A5049095056 @default.
• W3201628275 hasAuthorship W3201628275A5091549352 @default.
• W3201628275 hasBestOaLocation W32016282751 @default.
• W3201628275 hasConcept C105795698 @default.
• W3201628275 hasConcept C108583219 @default.
• W3201628275 hasConcept C114614502 @default.
• W3201628275 hasConcept C119857082 @default.
• W3201628275 hasConcept C147168706 @default.
• W3201628275 hasConcept C154945302 @default.
• W3201628275 hasConcept C155032097 @default.
• W3201628275 hasConcept C179518139 @default.
• W3201628275 hasConcept C2776214188 @default.
• W3201628275 hasConcept C2778061373 @default.
• W3201628275 hasConcept C28225019 @default.
• W3201628275 hasConcept C33923547 @default.
• W3201628275 hasConcept C41008148 @default.
• W3201628275 hasConcept C50644808 @default.
• W3201628275 hasConcept C60908668 @default.
• W3201628275 hasConcept C8521452 @default.
• W3201628275 hasConceptScore W3201628275C105795698 @default.
• W3201628275 hasConceptScore W3201628275C108583219 @default.
• W3201628275 hasConceptScore W3201628275C114614502 @default.
• W3201628275 hasConceptScore W3201628275C119857082 @default.
• W3201628275 hasConceptScore W3201628275C147168706 @default.
• W3201628275 hasConceptScore W3201628275C154945302 @default.
• W3201628275 hasConceptScore W3201628275C155032097 @default.
• W3201628275 hasConceptScore W3201628275C179518139 @default.
• W3201628275 hasConceptScore W3201628275C2776214188 @default.
• W3201628275 hasConceptScore W3201628275C2778061373 @default.
• W3201628275 hasConceptScore W3201628275C28225019 @default.
• W3201628275 hasConceptScore W3201628275C33923547 @default.
• W3201628275 hasConceptScore W3201628275C41008148 @default.
• W3201628275 hasConceptScore W3201628275C50644808 @default.
• W3201628275 hasConceptScore W3201628275C60908668 @default.
• W3201628275 hasConceptScore W3201628275C8521452 @default.
• W3201628275 hasIssue "7" @default.
• W3201628275 hasLocation W32016282751 @default.
• W3201628275 hasLocation W32016282752 @default.
• W3201628275 hasLocation W32016282753 @default.
• W3201628275 hasLocation W32016282754 @default.
• W3201628275 hasOpenAccess W3201628275 @default.
• W3201628275 hasPrimaryLocation W32016282751 @default.
• W3201628275 hasRelatedWork W1501213224 @default.
• W3201628275 hasRelatedWork W2480330148 @default.
• W3201628275 hasRelatedWork W4223943233 @default.
• W3201628275 hasRelatedWork W4224214554 @default.
• W3201628275 hasRelatedWork W4231994957 @default.
• W3201628275 hasRelatedWork W4312200629 @default.
• W3201628275 hasRelatedWork W4322750901 @default.

• next