Matches in SemOpenAlex for { <https://semopenalex.org/work/W3131020233> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W3131020233 abstract "Backpropagation of error (BP) is a widely used and highly successful learning algorithm. However, its reliance on non-local information in propagating error gradients makes it seem an unlikely candidate for learning in the brain. In the last decade, a number of investigations have been carried out focused upon determining whether alternative more biologically plausible computations can be used to approximate BP. This work builds on such a local learning algorithm - Gradient Adjusted Incremental Target Propagation (GAIT-prop) - which has recently been shown to approximate BP in a manner which appears biologically plausible. This method constructs local, layer-wise weight update targets in order to enable plausible credit assignment. However, in deep networks, the local weight updates computed by GAIT-prop can deviate from BP for a number of reasons. Here, we provide and test methods to overcome such sources of error. In particular, we adaptively rescale the locally-computed errors and show that this significantly increases the performance and stability of the GAIT-prop algorithm when applied to the CIFAR-10 dataset." @default.
- W3131020233 created "2021-03-01" @default.
- W3131020233 creator A5039391126 @default.
- W3131020233 creator A5054638114 @default.
- W3131020233 creator A5064518425 @default.
- W3131020233 creator A5087634622 @default.
- W3131020233 date "2021-02-23" @default.
- W3131020233 modified "2023-09-27" @default.
- W3131020233 title "Scaling up learning with GAIT-prop." @default.
- W3131020233 cites W1606458877 @default.
- W3131020233 cites W2076063813 @default.
- W3131020233 cites W2766736793 @default.
- W3131020233 cites W2919115771 @default.
- W3131020233 cites W2952448932 @default.
- W3131020233 cites W2964121744 @default.
- W3131020233 cites W3016391357 @default.
- W3131020233 cites W3037315641 @default.
- W3131020233 cites W3045675148 @default.
- W3131020233 cites W3118608800 @default.
- W3131020233 hasPublicationYear "2021" @default.
- W3131020233 type Work @default.
- W3131020233 sameAs 3131020233 @default.
- W3131020233 citedByCount "1" @default.
- W3131020233 countsByYear W31310202332021 @default.
- W3131020233 crossrefType "posted-content" @default.
- W3131020233 hasAuthorship W3131020233A5039391126 @default.
- W3131020233 hasAuthorship W3131020233A5054638114 @default.
- W3131020233 hasAuthorship W3131020233A5064518425 @default.
- W3131020233 hasAuthorship W3131020233A5087634622 @default.
- W3131020233 hasConcept C112972136 @default.
- W3131020233 hasConcept C11413529 @default.
- W3131020233 hasConcept C119857082 @default.
- W3131020233 hasConcept C151800584 @default.
- W3131020233 hasConcept C153180895 @default.
- W3131020233 hasConcept C154945302 @default.
- W3131020233 hasConcept C155032097 @default.
- W3131020233 hasConcept C2524010 @default.
- W3131020233 hasConcept C33923547 @default.
- W3131020233 hasConcept C41008148 @default.
- W3131020233 hasConcept C42407357 @default.
- W3131020233 hasConcept C45374587 @default.
- W3131020233 hasConcept C50644808 @default.
- W3131020233 hasConcept C86803240 @default.
- W3131020233 hasConcept C99844830 @default.
- W3131020233 hasConceptScore W3131020233C112972136 @default.
- W3131020233 hasConceptScore W3131020233C11413529 @default.
- W3131020233 hasConceptScore W3131020233C119857082 @default.
- W3131020233 hasConceptScore W3131020233C151800584 @default.
- W3131020233 hasConceptScore W3131020233C153180895 @default.
- W3131020233 hasConceptScore W3131020233C154945302 @default.
- W3131020233 hasConceptScore W3131020233C155032097 @default.
- W3131020233 hasConceptScore W3131020233C2524010 @default.
- W3131020233 hasConceptScore W3131020233C33923547 @default.
- W3131020233 hasConceptScore W3131020233C41008148 @default.
- W3131020233 hasConceptScore W3131020233C42407357 @default.
- W3131020233 hasConceptScore W3131020233C45374587 @default.
- W3131020233 hasConceptScore W3131020233C50644808 @default.
- W3131020233 hasConceptScore W3131020233C86803240 @default.
- W3131020233 hasConceptScore W3131020233C99844830 @default.
- W3131020233 hasLocation W31310202331 @default.
- W3131020233 hasOpenAccess W3131020233 @default.
- W3131020233 hasPrimaryLocation W31310202331 @default.
- W3131020233 hasRelatedWork W122801842 @default.
- W3131020233 hasRelatedWork W1569549387 @default.
- W3131020233 hasRelatedWork W2182779517 @default.
- W3131020233 hasRelatedWork W2182982932 @default.
- W3131020233 hasRelatedWork W2188519809 @default.
- W3131020233 hasRelatedWork W2769340754 @default.
- W3131020233 hasRelatedWork W2886907189 @default.
- W3131020233 hasRelatedWork W2890311316 @default.
- W3131020233 hasRelatedWork W2912648029 @default.
- W3131020233 hasRelatedWork W2914317573 @default.
- W3131020233 hasRelatedWork W2951388117 @default.
- W3131020233 hasRelatedWork W2951401169 @default.
- W3131020233 hasRelatedWork W2962998014 @default.
- W3131020233 hasRelatedWork W2977803677 @default.
- W3131020233 hasRelatedWork W3013013274 @default.
- W3131020233 hasRelatedWork W3033103365 @default.
- W3131020233 hasRelatedWork W3034464575 @default.
- W3131020233 hasRelatedWork W3104752551 @default.
- W3131020233 hasRelatedWork W3110869667 @default.
- W3131020233 hasRelatedWork W3203471143 @default.
- W3131020233 isParatext "false" @default.
- W3131020233 isRetracted "false" @default.
- W3131020233 magId "3131020233" @default.
- W3131020233 workType "article" @default.