Increase attractor capacity using an ensembled neural network

作者：

Highlights：

• Ensemble of diluted Attractor Neural Networks for pattern retrieval.

• Increase of network storage capacity by divide-and-conquer approach of subnetworks.

• Ensemble system triples maximal capacity of the single network with same wiring cost.

• Engineering application to limited memory systems: embedded systems or smartphones.

摘要

•Ensemble of diluted Attractor Neural Networks for pattern retrieval.•Increase of network storage capacity by divide-and-conquer approach of subnetworks.•Ensemble system triples maximal capacity of the single network with same wiring cost.•Engineering application to limited memory systems: embedded systems or smartphones.

论文关键词：Hopfield-type network,Synaptic dilution,Network wiring cost,Storage capacity,Ensemble of Attractor Neural Networks,Divide-and-conquer parallelizing

论文评审过程：Received 14 July 2016, Revised 22 November 2016, Accepted 24 November 2016, Available online 25 November 2016, Version of Record 1 December 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.11.035