edoc

Strategies to associate memories by unsupervised learning in neural networks

Agnes, Everton J. and Mizusaki, Beatriz E. P. and Erichsen Jr, Rubem and Brunnet, Leonardo G.. (2013) Strategies to associate memories by unsupervised learning in neural networks. AIP Conference Proceedings, 1510 (1). pp. 255-257.

Full text not available from this repository.

Official URL: https://edoc.unibas.ch/79133/

Downloads: Statistics Overview

Abstract

In this work we study the effects of three different strategies to associate memories in a neural network composed by both excitatory and inhibitory spiking neurons, which are randomly connected through recurrent excitatory and inhibitory synapses. The system is intended to store a number of memories, associated to spatial external inputs. The strategies consist in the presentation of the input patterns through trials in: i) ordered sequence; ii) random sequence; iii) clustered sequences. In addition, an order parameter indicating the correlation between the trials' activities is introduced to compute associative memory capacities and the quality of memory retrieval.
Faculties and Departments:05 Faculty of Science > Departement Biozentrum > Neurobiology
05 Faculty of Science > Departement Biozentrum > Neurobiology > Theoretical and computational neuroscience (Agnes)
UniBasel Contributors:Agnes, Everton Joao
Item Type:Article, refereed
Article Subtype:Research Article
Publisher:AIP Publishing
ISSN:0094-243X
e-ISSN:1551-7616
Note:Publication type according to Uni Basel Research Database: Journal article
Related URLs:
Identification Number:
Last Modified:13 May 2021 03:10
Deposited On:13 Nov 2020 08:36

Repository Staff Only: item control page