Matches in SemOpenAlex for { <https://semopenalex.org/work/W2786738752> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2786738752 endingPage "9378" @default.
- W2786738752 startingPage "9368" @default.
- W2786738752 abstract "The backpropagation of error algorithm (BP) is impossible to implement in a real brain. The recent success of deep networks in machine learning and AI, however, has inspired proposals for understanding how the brain might learn across multiple layers, and hence how it might approximate BP. As of yet, none of these proposals have been rigorously evaluated on tasks where BP-guided deep learning has proved critical, or in architectures more structured than simple fully-connected networks. Here we present results on scaling up biologically motivated models of deep learning on datasets which need deep networks with appropriate architectures to achieve good performance. We present results on the MNIST, CIFAR-10, and ImageNet datasets and explore variants of target-propagation (TP) and feedback alignment (FA) algorithms, and explore performance in both fully- and locally-connected architectures. We also introduce weight-transport-free variants of difference target propagation (DTP) modified to remove backpropagation from the penultimate layer. Many of these algorithms perform well for MNIST, but for CIFAR and ImageNet we find that TP and FA variants perform significantly worse than BP, especially for networks composed of locally connected units, opening questions about whether new architectures and algorithms are required to scale these approaches. Our results and implementation details help establish baselines for biologically motivated deep learning schemes going forward." @default.
- W2786738752 created "2018-02-23" @default.
- W2786738752 creator A5007668479 @default.
- W2786738752 creator A5024209719 @default.
- W2786738752 creator A5051039278 @default.
- W2786738752 creator A5066294254 @default.
- W2786738752 creator A5076043313 @default.
- W2786738752 creator A5089917436 @default.
- W2786738752 date "2018-06-15" @default.
- W2786738752 modified "2023-09-29" @default.
- W2786738752 title "Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures" @default.
- W2786738752 hasPublicationYear "2018" @default.
- W2786738752 type Work @default.
- W2786738752 sameAs 2786738752 @default.
- W2786738752 citedByCount "47" @default.
- W2786738752 countsByYear W27867387522018 @default.
- W2786738752 countsByYear W27867387522019 @default.
- W2786738752 countsByYear W27867387522020 @default.
- W2786738752 countsByYear W27867387522021 @default.
- W2786738752 crossrefType "proceedings-article" @default.
- W2786738752 hasAuthorship W2786738752A5007668479 @default.
- W2786738752 hasAuthorship W2786738752A5024209719 @default.
- W2786738752 hasAuthorship W2786738752A5051039278 @default.
- W2786738752 hasAuthorship W2786738752A5066294254 @default.
- W2786738752 hasAuthorship W2786738752A5076043313 @default.
- W2786738752 hasAuthorship W2786738752A5089917436 @default.
- W2786738752 hasConcept C108583219 @default.
- W2786738752 hasConcept C114614502 @default.
- W2786738752 hasConcept C119857082 @default.
- W2786738752 hasConcept C154945302 @default.
- W2786738752 hasConcept C155032097 @default.
- W2786738752 hasConcept C190502265 @default.
- W2786738752 hasConcept C28225019 @default.
- W2786738752 hasConcept C33923547 @default.
- W2786738752 hasConcept C41008148 @default.
- W2786738752 hasConcept C48044578 @default.
- W2786738752 hasConcept C50644808 @default.
- W2786738752 hasConcept C77088390 @default.
- W2786738752 hasConceptScore W2786738752C108583219 @default.
- W2786738752 hasConceptScore W2786738752C114614502 @default.
- W2786738752 hasConceptScore W2786738752C119857082 @default.
- W2786738752 hasConceptScore W2786738752C154945302 @default.
- W2786738752 hasConceptScore W2786738752C155032097 @default.
- W2786738752 hasConceptScore W2786738752C190502265 @default.
- W2786738752 hasConceptScore W2786738752C28225019 @default.
- W2786738752 hasConceptScore W2786738752C33923547 @default.
- W2786738752 hasConceptScore W2786738752C41008148 @default.
- W2786738752 hasConceptScore W2786738752C48044578 @default.
- W2786738752 hasConceptScore W2786738752C50644808 @default.
- W2786738752 hasConceptScore W2786738752C77088390 @default.
- W2786738752 hasLocation W27867387521 @default.
- W2786738752 hasOpenAccess W2786738752 @default.
- W2786738752 hasPrimaryLocation W27867387521 @default.
- W2786738752 hasRelatedWork W1498436455 @default.
- W2786738752 hasRelatedWork W1606458877 @default.
- W2786738752 hasRelatedWork W1777649940 @default.
- W2786738752 hasRelatedWork W1855112655 @default.
- W2786738752 hasRelatedWork W2003357516 @default.
- W2786738752 hasRelatedWork W2159110831 @default.
- W2786738752 hasRelatedWork W2194775991 @default.
- W2786738752 hasRelatedWork W2527798464 @default.
- W2786738752 hasRelatedWork W2529004582 @default.
- W2786738752 hasRelatedWork W2552737632 @default.
- W2786738752 hasRelatedWork W2903723132 @default.
- W2786738752 hasRelatedWork W2919115771 @default.
- W2786738752 hasRelatedWork W2949117887 @default.
- W2786738752 hasRelatedWork W2952448932 @default.
- W2786738752 hasRelatedWork W2963440770 @default.
- W2786738752 hasRelatedWork W2964115671 @default.
- W2786738752 hasRelatedWork W2964319207 @default.
- W2786738752 hasRelatedWork W2978368159 @default.
- W2786738752 hasRelatedWork W3016391357 @default.
- W2786738752 hasRelatedWork W3118608800 @default.
- W2786738752 hasVolume "31" @default.
- W2786738752 isParatext "false" @default.
- W2786738752 isRetracted "false" @default.
- W2786738752 magId "2786738752" @default.
- W2786738752 workType "article" @default.