Localized activity profiles and storage capacity of rate-based autoassociative networks

DOI10.1103/PHYSREVE.73.061904zbMATH Open1244.92027arXivcond-mat/0505349OpenAlexW2038082961WikidataQ51935425 ScholiaQ51935425MaRDI QIDQ2903683FDOQ2903683

Authors: Yasser Roudi, Alessandro Treves 

Publication date: 12 August 2012

Published in: Physical Review E. Third Series (Search for Journal in Brave)

Abstract: We study analytically the effect of metrically structured connectivity on the behavior of autoassociative networks. We focus on three simple rate-based model neurons: threshold-linear, binary or smoothly saturating units. For a connectivity which is short range enough the threshold-linear network shows localized retrieval states. The saturating and binary models also exhibit spatially modulated retrieval states if the highest activity level that they can achieve is above the maximum activity of the units in the stored patterns. In the zero quenched noise limit, we derive an analytical formula for the critical value of the connectivity width below which one observes spatially non-uniform retrieval states. Localization reduces storage capacity, but only by a factor of 2~3. The approach that we present here is generic in the sense that there are no specific assumptions on the single unit input-output function nor on the exact connectivity structure.

Full work available at URL: https://arxiv.org/abs/cond-mat/0505349

Recommendations

• An associative network with spatially organized connectivity
• What determines the capacity of autoassociative memories in the brain?
• Information capacity in recurrent McCulloch–Pitts networks with sparsely coded memory states
• scientific article; zbMATH DE number 1164156
• Short- and long-range connections in autoassociative memory

zbMATH Keywords

metrical connectivity

Mathematics Subject Classification ID

Systems biology, networks (92C42) Neural biology (92C20)

Cites Work

• Statistical Physics of Spin Glasses and Information Processing
• Modeling Brain Function
• Self-consistent signal-to-noise analysis and its application to analogue neural networks with asymmetric connections
• What determines the capacity of autoassociative memories in the brain?
• Computing and Stability in Cortical Networks
• Activity dynamics and propagation of synchronous spiking in locally connected random net\-works
• Quantitative study of attractor neural networks retrieving at low spike rates: II. Low-rate retrieval in symmetric networks
• The signal-to-noise analysis of the Little–Hopfield model revisited

Cited In (4)

• Short- and long-range connections in autoassociative memory
• Block information and topology in memory networks
• Imposing Biological Constraints onto an Abstract Neocortical Attractor Network Model
• Storing structured sparse memories in a multi-modular cortical network model

This page was built for publication: Localized activity profiles and storage capacity of rate-based autoassociative networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2903683)