Neural reconstruction with approximate message passing (NeuRAMP)

Alyson K. Fletcher, Sundeep Rangan, Lav R. Varshney, Aniruddha Bhargava

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

Abstract

Many functional descriptions of spiking neurons assume a cascade structure where inputs are passed through an initial linear filtering stage that produces a low-dimensional signal that drives subsequent nonlinear stages. This paper presents a novel and systematic parameter estimation procedure for such models and applies the method to two neural estimation problems: (i) compressed-sensing based neural mapping from multi-neuron excitation, and (ii) estimation of neural receptive fields in sensory neurons. The proposed estimation algorithm models the neurons via a graphical model and then estimates the parameters in the model using a recently-developed generalized approximate message passing (GAMP) method. The GAMP method is based on Gaussian approximations of loopy belief propagation. In the neural connectivity problem, the GAMP-based method is shown to be computational efficient, provides a more exact modeling of the sparsity, can incorporate nonlinearities in the output and significantly outperforms previous compressed-sensing methods. For the receptive field estimation, the GAMP method can also exploit inherent structured sparsity in the linear weights. The method is validated on estimation of linear nonlinear Poisson (LNP) cascade models for receptive fields of salamander retinal ganglion cells.

Original languageEnglish (US)
Title of host publicationAdvances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011
StatePublished - 2011
Event25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011 - Granada, Spain
Duration: Dec 12 2011Dec 14 2011

Other

Other25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011
CountrySpain
CityGranada
Period12/12/1112/14/11

Fingerprint

Message passing
Neurons
Compressed sensing
Parameter estimation

ASJC Scopus subject areas

  • Information Systems

Cite this

Fletcher, A. K., Rangan, S., Varshney, L. R., & Bhargava, A. (2011). Neural reconstruction with approximate message passing (NeuRAMP). In Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011

Neural reconstruction with approximate message passing (NeuRAMP). / Fletcher, Alyson K.; Rangan, Sundeep; Varshney, Lav R.; Bhargava, Aniruddha.

Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011. 2011.

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

Fletcher, AK, Rangan, S, Varshney, LR & Bhargava, A 2011, Neural reconstruction with approximate message passing (NeuRAMP). in Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011. 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011, Granada, Spain, 12/12/11.
Fletcher AK, Rangan S, Varshney LR, Bhargava A. Neural reconstruction with approximate message passing (NeuRAMP). In Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011. 2011
Fletcher, Alyson K. ; Rangan, Sundeep ; Varshney, Lav R. ; Bhargava, Aniruddha. / Neural reconstruction with approximate message passing (NeuRAMP). Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011. 2011.
@inproceedings{a699e541789c4f919fef43d9b515d9dc,
title = "Neural reconstruction with approximate message passing (NeuRAMP)",
abstract = "Many functional descriptions of spiking neurons assume a cascade structure where inputs are passed through an initial linear filtering stage that produces a low-dimensional signal that drives subsequent nonlinear stages. This paper presents a novel and systematic parameter estimation procedure for such models and applies the method to two neural estimation problems: (i) compressed-sensing based neural mapping from multi-neuron excitation, and (ii) estimation of neural receptive fields in sensory neurons. The proposed estimation algorithm models the neurons via a graphical model and then estimates the parameters in the model using a recently-developed generalized approximate message passing (GAMP) method. The GAMP method is based on Gaussian approximations of loopy belief propagation. In the neural connectivity problem, the GAMP-based method is shown to be computational efficient, provides a more exact modeling of the sparsity, can incorporate nonlinearities in the output and significantly outperforms previous compressed-sensing methods. For the receptive field estimation, the GAMP method can also exploit inherent structured sparsity in the linear weights. The method is validated on estimation of linear nonlinear Poisson (LNP) cascade models for receptive fields of salamander retinal ganglion cells.",
author = "Fletcher, {Alyson K.} and Sundeep Rangan and Varshney, {Lav R.} and Aniruddha Bhargava",
year = "2011",
language = "English (US)",
isbn = "9781618395993",
booktitle = "Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011",

}

TY - GEN

T1 - Neural reconstruction with approximate message passing (NeuRAMP)

AU - Fletcher, Alyson K.

AU - Rangan, Sundeep

AU - Varshney, Lav R.

AU - Bhargava, Aniruddha

PY - 2011

Y1 - 2011

N2 - Many functional descriptions of spiking neurons assume a cascade structure where inputs are passed through an initial linear filtering stage that produces a low-dimensional signal that drives subsequent nonlinear stages. This paper presents a novel and systematic parameter estimation procedure for such models and applies the method to two neural estimation problems: (i) compressed-sensing based neural mapping from multi-neuron excitation, and (ii) estimation of neural receptive fields in sensory neurons. The proposed estimation algorithm models the neurons via a graphical model and then estimates the parameters in the model using a recently-developed generalized approximate message passing (GAMP) method. The GAMP method is based on Gaussian approximations of loopy belief propagation. In the neural connectivity problem, the GAMP-based method is shown to be computational efficient, provides a more exact modeling of the sparsity, can incorporate nonlinearities in the output and significantly outperforms previous compressed-sensing methods. For the receptive field estimation, the GAMP method can also exploit inherent structured sparsity in the linear weights. The method is validated on estimation of linear nonlinear Poisson (LNP) cascade models for receptive fields of salamander retinal ganglion cells.

AB - Many functional descriptions of spiking neurons assume a cascade structure where inputs are passed through an initial linear filtering stage that produces a low-dimensional signal that drives subsequent nonlinear stages. This paper presents a novel and systematic parameter estimation procedure for such models and applies the method to two neural estimation problems: (i) compressed-sensing based neural mapping from multi-neuron excitation, and (ii) estimation of neural receptive fields in sensory neurons. The proposed estimation algorithm models the neurons via a graphical model and then estimates the parameters in the model using a recently-developed generalized approximate message passing (GAMP) method. The GAMP method is based on Gaussian approximations of loopy belief propagation. In the neural connectivity problem, the GAMP-based method is shown to be computational efficient, provides a more exact modeling of the sparsity, can incorporate nonlinearities in the output and significantly outperforms previous compressed-sensing methods. For the receptive field estimation, the GAMP method can also exploit inherent structured sparsity in the linear weights. The method is validated on estimation of linear nonlinear Poisson (LNP) cascade models for receptive fields of salamander retinal ganglion cells.

UR - http://www.scopus.com/inward/record.url?scp=84860642525&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860642525&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781618395993

BT - Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011, NIPS 2011

ER -