Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384648110> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4384648110 abstract "We study bi-directional associative neural networks that, exposed to noisy examples of an extensive number of random archetypes, learn the latter (with or without the presence of a teacher) when the supplied information is enough: in this setting, learning is heteroassociative -- involving couples of patterns -- and it is achieved by reverberating the information depicted from the examples through the layers of the network. By adapting Guerra's interpolation technique, we provide a full statistical mechanical picture of supervised and unsupervised learning processes (at the replica symmetric level of description) obtaining analytically phase diagrams, thresholds for learning, a picture of the ground-state in plain agreement with Monte Carlo simulations and signal-to-noise outcomes. In the large dataset limit, the Kosko storage prescription as well as its statistical mechanical picture provided by Kurchan, Peliti, and Saber in the eighties is fully recovered. Computational advantages in dealing with information reverberation, rather than storage, are discussed for natural test cases. In particular, we show how this network admits an integral representation in terms of two coupled restricted Boltzmann machines, whose hidden layers are entirely built of by grand-mother neurons, to prove that by coupling solely these grand-mother neurons we can correlate the patterns they are related to: it is thus possible to recover Pavlov's Classical Conditioning by adding just one synapse among the correct grand-mother neurons (hence saving an extensive number of these links for further information storage w.r.t. the classical autoassociative setting)." @default.
- W4384648110 created "2023-07-19" @default.
- W4384648110 creator A5011672584 @default.
- W4384648110 creator A5045686781 @default.
- W4384648110 creator A5048983044 @default.
- W4384648110 date "2023-07-17" @default.
- W4384648110 modified "2023-10-18" @default.
- W4384648110 title "Statistical Mechanics of Learning via Reverberation in Bidirectional Associative Memories" @default.
- W4384648110 doi "https://doi.org/10.48550/arxiv.2307.08365" @default.
- W4384648110 hasPublicationYear "2023" @default.
- W4384648110 type Work @default.
- W4384648110 citedByCount "0" @default.
- W4384648110 crossrefType "posted-content" @default.
- W4384648110 hasAuthorship W4384648110A5011672584 @default.
- W4384648110 hasAuthorship W4384648110A5045686781 @default.
- W4384648110 hasAuthorship W4384648110A5048983044 @default.
- W4384648110 hasBestOaLocation W43846481101 @default.
- W4384648110 hasConcept C115961682 @default.
- W4384648110 hasConcept C121332964 @default.
- W4384648110 hasConcept C121864883 @default.
- W4384648110 hasConcept C127413603 @default.
- W4384648110 hasConcept C131584629 @default.
- W4384648110 hasConcept C154945302 @default.
- W4384648110 hasConcept C17744445 @default.
- W4384648110 hasConcept C192576344 @default.
- W4384648110 hasConcept C199354608 @default.
- W4384648110 hasConcept C199539241 @default.
- W4384648110 hasConcept C24890656 @default.
- W4384648110 hasConcept C2776359362 @default.
- W4384648110 hasConcept C41008148 @default.
- W4384648110 hasConcept C50644808 @default.
- W4384648110 hasConcept C53442348 @default.
- W4384648110 hasConcept C78519656 @default.
- W4384648110 hasConcept C8038995 @default.
- W4384648110 hasConcept C80444323 @default.
- W4384648110 hasConcept C94625758 @default.
- W4384648110 hasConcept C95851461 @default.
- W4384648110 hasConcept C99498987 @default.
- W4384648110 hasConcept C99874945 @default.
- W4384648110 hasConceptScore W4384648110C115961682 @default.
- W4384648110 hasConceptScore W4384648110C121332964 @default.
- W4384648110 hasConceptScore W4384648110C121864883 @default.
- W4384648110 hasConceptScore W4384648110C127413603 @default.
- W4384648110 hasConceptScore W4384648110C131584629 @default.
- W4384648110 hasConceptScore W4384648110C154945302 @default.
- W4384648110 hasConceptScore W4384648110C17744445 @default.
- W4384648110 hasConceptScore W4384648110C192576344 @default.
- W4384648110 hasConceptScore W4384648110C199354608 @default.
- W4384648110 hasConceptScore W4384648110C199539241 @default.
- W4384648110 hasConceptScore W4384648110C24890656 @default.
- W4384648110 hasConceptScore W4384648110C2776359362 @default.
- W4384648110 hasConceptScore W4384648110C41008148 @default.
- W4384648110 hasConceptScore W4384648110C50644808 @default.
- W4384648110 hasConceptScore W4384648110C53442348 @default.
- W4384648110 hasConceptScore W4384648110C78519656 @default.
- W4384648110 hasConceptScore W4384648110C8038995 @default.
- W4384648110 hasConceptScore W4384648110C80444323 @default.
- W4384648110 hasConceptScore W4384648110C94625758 @default.
- W4384648110 hasConceptScore W4384648110C95851461 @default.
- W4384648110 hasConceptScore W4384648110C99498987 @default.
- W4384648110 hasConceptScore W4384648110C99874945 @default.
- W4384648110 hasLocation W43846481101 @default.
- W4384648110 hasOpenAccess W4384648110 @default.
- W4384648110 hasPrimaryLocation W43846481101 @default.
- W4384648110 hasRelatedWork W2047776774 @default.
- W4384648110 hasRelatedWork W2105865682 @default.
- W4384648110 hasRelatedWork W2607300641 @default.
- W4384648110 hasRelatedWork W2956138382 @default.
- W4384648110 hasRelatedWork W2971852391 @default.
- W4384648110 hasRelatedWork W3044458868 @default.
- W4384648110 hasRelatedWork W3082895349 @default.
- W4384648110 hasRelatedWork W3104798141 @default.
- W4384648110 hasRelatedWork W3116923736 @default.
- W4384648110 hasRelatedWork W3123344745 @default.
- W4384648110 isParatext "false" @default.
- W4384648110 isRetracted "false" @default.
- W4384648110 workType "article" @default.