Independent component analysis in spiking neurons

Cristina Savin, Prashant Joshi, Jochen Triesch

Research output: Contribution to journalArticle

Abstract

Although models based on independent component analysis (ICA) have been successful in explaining various properties of sensory coding in the cortex, it remains unclear how networks of spiking neurons using realistic plasticity rules can realize such computation. Here, we propose a biologically plausible mechanism for ICA-like learning with spiking neurons. Our model combines spike-timing dependent plasticity and synaptic scaling with an intrinsic plasticity rule that regulates neuronal excitability to maximize information transmission. We show that a stochastically spiking neuron learns one independent component for inputs encoded either as rates or using spike-spike correlations. Furthermore, different independent components can be recovered, when the activity of different neurons is decorrelated by adaptive lateral inhibition.

Original languageEnglish (US)
JournalPLoS Computational Biology
Volume6
Issue number4
DOIs
StatePublished - Apr 2010

Fingerprint

Spiking Neurons
Independent component analysis
Independent Component Analysis
Spike
Plasticity
Neurons
plasticity
neurons
Excitability
Cortex
Neuronal Plasticity
Neuron
Timing
Lateral
Coding
learning
Maximise
Scaling
Model-based
sensory properties

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Modeling and Simulation
  • Ecology
  • Molecular Biology
  • Genetics
  • Cellular and Molecular Neuroscience
  • Computational Theory and Mathematics

Cite this

Independent component analysis in spiking neurons. / Savin, Cristina; Joshi, Prashant; Triesch, Jochen.

In: PLoS Computational Biology, Vol. 6, No. 4, 04.2010.

Research output: Contribution to journalArticle

Savin, Cristina ; Joshi, Prashant ; Triesch, Jochen. / Independent component analysis in spiking neurons. In: PLoS Computational Biology. 2010 ; Vol. 6, No. 4.
@article{34cbc5c1f74b43dc905c5a204145e1b3,
title = "Independent component analysis in spiking neurons",
abstract = "Although models based on independent component analysis (ICA) have been successful in explaining various properties of sensory coding in the cortex, it remains unclear how networks of spiking neurons using realistic plasticity rules can realize such computation. Here, we propose a biologically plausible mechanism for ICA-like learning with spiking neurons. Our model combines spike-timing dependent plasticity and synaptic scaling with an intrinsic plasticity rule that regulates neuronal excitability to maximize information transmission. We show that a stochastically spiking neuron learns one independent component for inputs encoded either as rates or using spike-spike correlations. Furthermore, different independent components can be recovered, when the activity of different neurons is decorrelated by adaptive lateral inhibition.",
author = "Cristina Savin and Prashant Joshi and Jochen Triesch",
year = "2010",
month = "4",
doi = "10.1371/journal.pcbi.1000757",
language = "English (US)",
volume = "6",
journal = "PLoS Computational Biology",
issn = "1553-734X",
publisher = "Public Library of Science",
number = "4",

}

TY - JOUR

T1 - Independent component analysis in spiking neurons

AU - Savin, Cristina

AU - Joshi, Prashant

AU - Triesch, Jochen

PY - 2010/4

Y1 - 2010/4

N2 - Although models based on independent component analysis (ICA) have been successful in explaining various properties of sensory coding in the cortex, it remains unclear how networks of spiking neurons using realistic plasticity rules can realize such computation. Here, we propose a biologically plausible mechanism for ICA-like learning with spiking neurons. Our model combines spike-timing dependent plasticity and synaptic scaling with an intrinsic plasticity rule that regulates neuronal excitability to maximize information transmission. We show that a stochastically spiking neuron learns one independent component for inputs encoded either as rates or using spike-spike correlations. Furthermore, different independent components can be recovered, when the activity of different neurons is decorrelated by adaptive lateral inhibition.

AB - Although models based on independent component analysis (ICA) have been successful in explaining various properties of sensory coding in the cortex, it remains unclear how networks of spiking neurons using realistic plasticity rules can realize such computation. Here, we propose a biologically plausible mechanism for ICA-like learning with spiking neurons. Our model combines spike-timing dependent plasticity and synaptic scaling with an intrinsic plasticity rule that regulates neuronal excitability to maximize information transmission. We show that a stochastically spiking neuron learns one independent component for inputs encoded either as rates or using spike-spike correlations. Furthermore, different independent components can be recovered, when the activity of different neurons is decorrelated by adaptive lateral inhibition.

UR - http://www.scopus.com/inward/record.url?scp=77954576671&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954576671&partnerID=8YFLogxK

U2 - 10.1371/journal.pcbi.1000757

DO - 10.1371/journal.pcbi.1000757

M3 - Article

VL - 6

JO - PLoS Computational Biology

JF - PLoS Computational Biology

SN - 1553-734X

IS - 4

ER -