Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313304716> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W4313304716 abstract "The aim of this paper is to introduce a new learning procedure for neural networks and to demonstrate that it works well enough on a few small problems to be worth further investigation. The Forward-Forward algorithm replaces the forward and backward passes of backpropagation by two forward passes, one with positive (i.e. real) data and the other with negative data which could be generated by the network itself. Each layer has its own objective function which is simply to have high goodness for positive data and low goodness for negative data. The sum of the squared activities in a layer can be used as the goodness but there are many other possibilities, including minus the sum of the squared activities. If the positive and negative passes could be separated in time, the negative passes could be done offline, which would make the learning much simpler in the positive pass and allow video to be pipelined through the network without ever storing activities or stopping to propagate derivatives." @default.
- W4313304716 created "2023-01-06" @default.
- W4313304716 creator A5024209719 @default.
- W4313304716 date "2022-12-26" @default.
- W4313304716 modified "2023-10-12" @default.
- W4313304716 title "The Forward-Forward Algorithm: Some Preliminary Investigations" @default.
- W4313304716 doi "https://doi.org/10.48550/arxiv.2212.13345" @default.
- W4313304716 hasPublicationYear "2022" @default.
- W4313304716 type Work @default.
- W4313304716 citedByCount "1" @default.
- W4313304716 crossrefType "posted-content" @default.
- W4313304716 hasAuthorship W4313304716A5024209719 @default.
- W4313304716 hasBestOaLocation W43133047161 @default.
- W4313304716 hasConcept C11413529 @default.
- W4313304716 hasConcept C119857082 @default.
- W4313304716 hasConcept C132480984 @default.
- W4313304716 hasConcept C14036430 @default.
- W4313304716 hasConcept C154945302 @default.
- W4313304716 hasConcept C155032097 @default.
- W4313304716 hasConcept C178790620 @default.
- W4313304716 hasConcept C185592680 @default.
- W4313304716 hasConcept C2779227376 @default.
- W4313304716 hasConcept C41008148 @default.
- W4313304716 hasConcept C50644808 @default.
- W4313304716 hasConcept C78458016 @default.
- W4313304716 hasConcept C86803240 @default.
- W4313304716 hasConceptScore W4313304716C11413529 @default.
- W4313304716 hasConceptScore W4313304716C119857082 @default.
- W4313304716 hasConceptScore W4313304716C132480984 @default.
- W4313304716 hasConceptScore W4313304716C14036430 @default.
- W4313304716 hasConceptScore W4313304716C154945302 @default.
- W4313304716 hasConceptScore W4313304716C155032097 @default.
- W4313304716 hasConceptScore W4313304716C178790620 @default.
- W4313304716 hasConceptScore W4313304716C185592680 @default.
- W4313304716 hasConceptScore W4313304716C2779227376 @default.
- W4313304716 hasConceptScore W4313304716C41008148 @default.
- W4313304716 hasConceptScore W4313304716C50644808 @default.
- W4313304716 hasConceptScore W4313304716C78458016 @default.
- W4313304716 hasConceptScore W4313304716C86803240 @default.
- W4313304716 hasLocation W43133047161 @default.
- W4313304716 hasOpenAccess W4313304716 @default.
- W4313304716 hasPrimaryLocation W43133047161 @default.
- W4313304716 hasRelatedWork W2105899231 @default.
- W4313304716 hasRelatedWork W2159443810 @default.
- W4313304716 hasRelatedWork W2348584318 @default.
- W4313304716 hasRelatedWork W2350832155 @default.
- W4313304716 hasRelatedWork W2362189222 @default.
- W4313304716 hasRelatedWork W2383491577 @default.
- W4313304716 hasRelatedWork W2391384657 @default.
- W4313304716 hasRelatedWork W2186629877 @default.
- W4313304716 hasRelatedWork W2516580779 @default.
- W4313304716 hasRelatedWork W2609564064 @default.
- W4313304716 isParatext "false" @default.
- W4313304716 isRetracted "false" @default.
- W4313304716 workType "article" @default.