Computational Modelling of a Retinal Ganglion Cells' Receptive Field using Machine Learning Approaches

Research output: Contribution to journalArticle

Abstract

Artificial vision using computational models that can mimic biological vision is an area of ongoing research. One of the main themes within this research is the study of the retina and in particular, retinal ganglion cells which are responsible for encoding the visual stimuli. A common approach to modelling the internal processes of retinal ganglion cells is the use of a linear - non-linear cascade model, which models the cell's response using a linear filter followed by a static non-linearity. However, the resulting model is generally restrictive as it is often a poor estimator of the neuron's response. In this paper we present an alternative to the linear - non-linear model by modelling retinal ganglion cells using a number of machine learning techniques which have a proven track record for learning complex non-linearities in many different domains. A comparison of the model predicted spike rate shows that the machine learning models perform better than the standard linear - non-linear approach in the case of temporal white noise stimuli.
LanguageEnglish
Number of pages28
JournalNeurocomputing
Early online date6 Oct 2018
DOIs
Publication statusE-pub ahead of print - 6 Oct 2018

Fingerprint

Retinal Ganglion Cells
Learning systems
Nonlinear Dynamics
Research
Retina
Learning
Neurons
White noise
Machine Learning
Computer vision

Keywords

  • Artificial Vision
  • Biological Vision
  • Machine Learning
  • Retinal Ganglion Cell
  • Receptive Field

Cite this

@article{bf9be5e12f6f41c692895856329bdf3a,
title = "Computational Modelling of a Retinal Ganglion Cells' Receptive Field using Machine Learning Approaches",
abstract = "Artificial vision using computational models that can mimic biological vision is an area of ongoing research. One of the main themes within this research is the study of the retina and in particular, retinal ganglion cells which are responsible for encoding the visual stimuli. A common approach to modelling the internal processes of retinal ganglion cells is the use of a linear - non-linear cascade model, which models the cell's response using a linear filter followed by a static non-linearity. However, the resulting model is generally restrictive as it is often a poor estimator of the neuron's response. In this paper we present an alternative to the linear - non-linear model by modelling retinal ganglion cells using a number of machine learning techniques which have a proven track record for learning complex non-linearities in many different domains. A comparison of the model predicted spike rate shows that the machine learning models perform better than the standard linear - non-linear approach in the case of temporal white noise stimuli.",
keywords = "Artificial Vision, Biological Vision, Machine Learning, Retinal Ganglion Cell, Receptive Field",
author = "Gautham Das and Philip Vance and Dermot Kerr and Sonya Coleman and T.Martin McGinnity and Jian Liu",
year = "2018",
month = "10",
day = "6",
doi = "10.1016/j.neucom.2018.10.004",
language = "English",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Computational Modelling of a Retinal Ganglion Cells' Receptive Field using Machine Learning Approaches

AU - Das, Gautham

AU - Vance, Philip

AU - Kerr, Dermot

AU - Coleman, Sonya

AU - McGinnity, T.Martin

AU - Liu, Jian

PY - 2018/10/6

Y1 - 2018/10/6

N2 - Artificial vision using computational models that can mimic biological vision is an area of ongoing research. One of the main themes within this research is the study of the retina and in particular, retinal ganglion cells which are responsible for encoding the visual stimuli. A common approach to modelling the internal processes of retinal ganglion cells is the use of a linear - non-linear cascade model, which models the cell's response using a linear filter followed by a static non-linearity. However, the resulting model is generally restrictive as it is often a poor estimator of the neuron's response. In this paper we present an alternative to the linear - non-linear model by modelling retinal ganglion cells using a number of machine learning techniques which have a proven track record for learning complex non-linearities in many different domains. A comparison of the model predicted spike rate shows that the machine learning models perform better than the standard linear - non-linear approach in the case of temporal white noise stimuli.

AB - Artificial vision using computational models that can mimic biological vision is an area of ongoing research. One of the main themes within this research is the study of the retina and in particular, retinal ganglion cells which are responsible for encoding the visual stimuli. A common approach to modelling the internal processes of retinal ganglion cells is the use of a linear - non-linear cascade model, which models the cell's response using a linear filter followed by a static non-linearity. However, the resulting model is generally restrictive as it is often a poor estimator of the neuron's response. In this paper we present an alternative to the linear - non-linear model by modelling retinal ganglion cells using a number of machine learning techniques which have a proven track record for learning complex non-linearities in many different domains. A comparison of the model predicted spike rate shows that the machine learning models perform better than the standard linear - non-linear approach in the case of temporal white noise stimuli.

KW - Artificial Vision

KW - Biological Vision

KW - Machine Learning

KW - Retinal Ganglion Cell

KW - Receptive Field

U2 - 10.1016/j.neucom.2018.10.004

DO - 10.1016/j.neucom.2018.10.004

M3 - Article

JO - Neurocomputing

T2 - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -