A robust biologically plausible implementation of ICA-like learning

Felipe Gerhard, Cristina Savin, Jochen Triesch

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a model that can perform ICA-like learning by simple, local, biologically plausible rules. By combining synaptic learning with homeostatic regulation of neuron properties and adaptive lateral inhibition, the neural network can robustly learn Gabor-like receptive fields from natural images. With spatially localized inhibitory connections, a topographic map can be achieved. Additionally, the network can solve the Földiák bars problem, a classical nonlinear ICA task.

Original languageEnglish (US)
Title of host publicationESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning
Pages147-152
Number of pages6
StatePublished - 2009
Event17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning, ESANN 2009 - Bruges, Belgium
Duration: Apr 22 2009Apr 24 2009

Other

Other17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning, ESANN 2009
CountryBelgium
CityBruges
Period4/22/094/24/09

Fingerprint

Independent component analysis
Neurons
Neural networks

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Cite this

Gerhard, F., Savin, C., & Triesch, J. (2009). A robust biologically plausible implementation of ICA-like learning. In ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning (pp. 147-152)

A robust biologically plausible implementation of ICA-like learning. / Gerhard, Felipe; Savin, Cristina; Triesch, Jochen.

ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning. 2009. p. 147-152.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gerhard, F, Savin, C & Triesch, J 2009, A robust biologically plausible implementation of ICA-like learning. in ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning. pp. 147-152, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning, ESANN 2009, Bruges, Belgium, 4/22/09.
Gerhard F, Savin C, Triesch J. A robust biologically plausible implementation of ICA-like learning. In ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning. 2009. p. 147-152
Gerhard, Felipe ; Savin, Cristina ; Triesch, Jochen. / A robust biologically plausible implementation of ICA-like learning. ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning. 2009. pp. 147-152
@inproceedings{98796d28e5144d3bbe9fc7a718b82b91,
title = "A robust biologically plausible implementation of ICA-like learning",
abstract = "We present a model that can perform ICA-like learning by simple, local, biologically plausible rules. By combining synaptic learning with homeostatic regulation of neuron properties and adaptive lateral inhibition, the neural network can robustly learn Gabor-like receptive fields from natural images. With spatially localized inhibitory connections, a topographic map can be achieved. Additionally, the network can solve the F{\"o}ldi{\'a}k bars problem, a classical nonlinear ICA task.",
author = "Felipe Gerhard and Cristina Savin and Jochen Triesch",
year = "2009",
language = "English (US)",
isbn = "2930307099",
pages = "147--152",
booktitle = "ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning",

}

TY - GEN

T1 - A robust biologically plausible implementation of ICA-like learning

AU - Gerhard, Felipe

AU - Savin, Cristina

AU - Triesch, Jochen

PY - 2009

Y1 - 2009

N2 - We present a model that can perform ICA-like learning by simple, local, biologically plausible rules. By combining synaptic learning with homeostatic regulation of neuron properties and adaptive lateral inhibition, the neural network can robustly learn Gabor-like receptive fields from natural images. With spatially localized inhibitory connections, a topographic map can be achieved. Additionally, the network can solve the Földiák bars problem, a classical nonlinear ICA task.

AB - We present a model that can perform ICA-like learning by simple, local, biologically plausible rules. By combining synaptic learning with homeostatic regulation of neuron properties and adaptive lateral inhibition, the neural network can robustly learn Gabor-like receptive fields from natural images. With spatially localized inhibitory connections, a topographic map can be achieved. Additionally, the network can solve the Földiák bars problem, a classical nonlinear ICA task.

UR - http://www.scopus.com/inward/record.url?scp=84887012006&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887012006&partnerID=8YFLogxK

M3 - Conference contribution

SN - 2930307099

SN - 9782930307091

SP - 147

EP - 152

BT - ESANN 2009 Proceedings, 17th European Symposium on Artificial Neural Networks - Advances in Computational Intelligence and Learning

ER -