English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We study bi-directional associative neural networks that, exposed to noisyexamples of an extensive number of random archetypes, learn the latter (with orwithout the presence of a teacher) when the supplied information is enough: inthis setting, learning is heteroassociative -- involving couples of patterns --and it is achieved by reverberating the information depicted from the examplesthrough the layers of the network. By adapting Guerra's interpolationtechnique, we provide a full statistical mechanical picture of supervised andunsupervised learning processes (at the replica symmetric level of description)obtaining analytically phase diagrams, thresholds for learning, a picture ofthe ground-state in plain agreement with Monte Carlo simulations andsignal-to-noise outcomes. In the large dataset limit, the Kosko storageprescription as well as its statistical mechanical picture provided by Kurchan,Peliti, and Saber in the eighties is fully recovered. Computational advantagesin dealing with information reverberation, rather than storage, are discussedfor natural test cases. In particular, we show how this network admits anintegral representation in terms of two coupled restricted Boltzmann machines,whose hidden layers are entirely built of by grand-mother neurons, to provethat by coupling solely these grand-mother neurons we can correlate thepatterns they are related to: it is thus possible to recover Pavlov's ClassicalConditioning by adding just one synapse among the correct grand-mother neurons(hence saving an extensive number of these links for further informationstorage w.r.t. the classical autoassociative setting).

Statistical Mechanics of Learning via Reverberation in Bidirectional Associative Memories