Matches in SemOpenAlex for { <https://semopenalex.org/work/W1967364564> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W1967364564 endingPage "916" @default.
- W1967364564 startingPage "902" @default.
- W1967364564 abstract "We consider the generalization problem for a perceptron with binary synapses, implementing the Stochastic Belief-Propagation-Inspired (SBPI) learning algorithm which we proposed earlier, and perform a mean-field calculation to obtain a differential equation which describes the behaviour of the device in the limit of a large number of synapses N. We show that the solving time of SBPI is of order $Nsqrt{log N}$ , while the similar, well-known clipped perceptron (CP) algorithm does not converge to a solution at all in the time frame we considered. The analysis gives some insight into the ongoing process and shows that, in this context, the SBPI algorithm is equivalent to a new, simpler algorithm, which only differs from the CP algorithm by the addition of a stochastic, unsupervised meta-plastic reinforcement process, whose rate of application must be less than $sqrt{2/(pi N)}$ for the learning to be achieved effectively. The analytical results are confirmed by simulations." @default.
- W1967364564 created "2016-06-24" @default.
- W1967364564 creator A5028083013 @default.
- W1967364564 date "2009-09-01" @default.
- W1967364564 modified "2023-09-26" @default.
- W1967364564 title "Generalization Learning in a Perceptron with Binary Synapses" @default.
- W1967364564 cites W1650706190 @default.
- W1967364564 cites W196871588 @default.
- W1967364564 cites W2002193713 @default.
- W1967364564 cites W2021937730 @default.
- W1967364564 cites W2033606703 @default.
- W1967364564 cites W2036759514 @default.
- W1967364564 cites W2050665571 @default.
- W1967364564 cites W2060941075 @default.
- W1967364564 cites W2083910902 @default.
- W1967364564 cites W2085138324 @default.
- W1967364564 cites W2096551329 @default.
- W1967364564 cites W2107524792 @default.
- W1967364564 cites W2129386380 @default.
- W1967364564 cites W2133667940 @default.
- W1967364564 cites W2170282164 @default.
- W1967364564 doi "https://doi.org/10.1007/s10955-009-9822-1" @default.
- W1967364564 hasPublicationYear "2009" @default.
- W1967364564 type Work @default.
- W1967364564 sameAs 1967364564 @default.
- W1967364564 citedByCount "20" @default.
- W1967364564 countsByYear W19673645642013 @default.
- W1967364564 countsByYear W19673645642015 @default.
- W1967364564 countsByYear W19673645642016 @default.
- W1967364564 countsByYear W19673645642017 @default.
- W1967364564 countsByYear W19673645642018 @default.
- W1967364564 countsByYear W19673645642019 @default.
- W1967364564 countsByYear W19673645642020 @default.
- W1967364564 countsByYear W19673645642021 @default.
- W1967364564 countsByYear W19673645642022 @default.
- W1967364564 countsByYear W19673645642023 @default.
- W1967364564 crossrefType "journal-article" @default.
- W1967364564 hasAuthorship W1967364564A5028083013 @default.
- W1967364564 hasBestOaLocation W19673645642 @default.
- W1967364564 hasConcept C105795698 @default.
- W1967364564 hasConcept C11413529 @default.
- W1967364564 hasConcept C134306372 @default.
- W1967364564 hasConcept C151201525 @default.
- W1967364564 hasConcept C151730666 @default.
- W1967364564 hasConcept C154945302 @default.
- W1967364564 hasConcept C177148314 @default.
- W1967364564 hasConcept C2779343474 @default.
- W1967364564 hasConcept C33923547 @default.
- W1967364564 hasConcept C41008148 @default.
- W1967364564 hasConcept C48372109 @default.
- W1967364564 hasConcept C50644808 @default.
- W1967364564 hasConcept C60908668 @default.
- W1967364564 hasConcept C8038995 @default.
- W1967364564 hasConcept C8272713 @default.
- W1967364564 hasConcept C86803240 @default.
- W1967364564 hasConcept C94375191 @default.
- W1967364564 hasConcept C97541855 @default.
- W1967364564 hasConceptScore W1967364564C105795698 @default.
- W1967364564 hasConceptScore W1967364564C11413529 @default.
- W1967364564 hasConceptScore W1967364564C134306372 @default.
- W1967364564 hasConceptScore W1967364564C151201525 @default.
- W1967364564 hasConceptScore W1967364564C151730666 @default.
- W1967364564 hasConceptScore W1967364564C154945302 @default.
- W1967364564 hasConceptScore W1967364564C177148314 @default.
- W1967364564 hasConceptScore W1967364564C2779343474 @default.
- W1967364564 hasConceptScore W1967364564C33923547 @default.
- W1967364564 hasConceptScore W1967364564C41008148 @default.
- W1967364564 hasConceptScore W1967364564C48372109 @default.
- W1967364564 hasConceptScore W1967364564C50644808 @default.
- W1967364564 hasConceptScore W1967364564C60908668 @default.
- W1967364564 hasConceptScore W1967364564C8038995 @default.
- W1967364564 hasConceptScore W1967364564C8272713 @default.
- W1967364564 hasConceptScore W1967364564C86803240 @default.
- W1967364564 hasConceptScore W1967364564C94375191 @default.
- W1967364564 hasConceptScore W1967364564C97541855 @default.
- W1967364564 hasIssue "5" @default.
- W1967364564 hasLocation W19673645641 @default.
- W1967364564 hasLocation W19673645642 @default.
- W1967364564 hasOpenAccess W1967364564 @default.
- W1967364564 hasPrimaryLocation W19673645641 @default.
- W1967364564 hasRelatedWork W1517945184 @default.
- W1967364564 hasRelatedWork W1967364564 @default.
- W1967364564 hasRelatedWork W2071201386 @default.
- W1967364564 hasRelatedWork W2123845559 @default.
- W1967364564 hasRelatedWork W2941320171 @default.
- W1967364564 hasRelatedWork W3022038857 @default.
- W1967364564 hasRelatedWork W3074294383 @default.
- W1967364564 hasRelatedWork W3208179741 @default.
- W1967364564 hasRelatedWork W4297022372 @default.
- W1967364564 hasRelatedWork W4377293004 @default.
- W1967364564 hasVolume "136" @default.
- W1967364564 isParatext "false" @default.
- W1967364564 isRetracted "false" @default.
- W1967364564 magId "1967364564" @default.
- W1967364564 workType "article" @default.