Normalization and pooling in hierarchical models of natural images

Luis G. Sanchez-Giraldo, Md Nasir Uddin Laskar, Odelia Schwartz

Research output: Contribution to journalReview article

Abstract

Divisive normalization and subunit pooling are two canonical classes of computation that have become widely used in descriptive (what) models of visual cortical processing. Normative (why) models from natural image statistics can help constrain the form and parameters of such classes of models. We focus on recent advances in two particular directions, namely deriving richer forms of divisive normalization, and advances in learning pooling from image statistics. We discuss the incorporation of such components into hierarchical models. We consider both hierarchical unsupervised learning from image statistics, and discriminative supervised learning in deep convolutional neural networks (CNNs). We further discuss studies on the utility and extensions of the convolutional architecture, which has also been adopted by recent descriptive models. We review the recent literature and discuss the current promises and gaps of using such approaches to gain a better understanding of how cortical neurons represent and process complex visual stimuli.

Original languageEnglish (US)
Pages (from-to)65-72
Number of pages8
JournalCurrent Opinion in Neurobiology
Volume55
DOIs
StatePublished - Apr 1 2019

Fingerprint

Learning
Neurons
Direction compound

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

Normalization and pooling in hierarchical models of natural images. / Sanchez-Giraldo, Luis G.; Laskar, Md Nasir Uddin; Schwartz, Odelia.

In: Current Opinion in Neurobiology, Vol. 55, 01.04.2019, p. 65-72.

Research output: Contribution to journalReview article

Sanchez-Giraldo, Luis G. ; Laskar, Md Nasir Uddin ; Schwartz, Odelia. / Normalization and pooling in hierarchical models of natural images. In: Current Opinion in Neurobiology. 2019 ; Vol. 55. pp. 65-72.
@article{a48229d073a44f9b82af22491d92d208,
title = "Normalization and pooling in hierarchical models of natural images",
abstract = "Divisive normalization and subunit pooling are two canonical classes of computation that have become widely used in descriptive (what) models of visual cortical processing. Normative (why) models from natural image statistics can help constrain the form and parameters of such classes of models. We focus on recent advances in two particular directions, namely deriving richer forms of divisive normalization, and advances in learning pooling from image statistics. We discuss the incorporation of such components into hierarchical models. We consider both hierarchical unsupervised learning from image statistics, and discriminative supervised learning in deep convolutional neural networks (CNNs). We further discuss studies on the utility and extensions of the convolutional architecture, which has also been adopted by recent descriptive models. We review the recent literature and discuss the current promises and gaps of using such approaches to gain a better understanding of how cortical neurons represent and process complex visual stimuli.",
author = "Sanchez-Giraldo, {Luis G.} and Laskar, {Md Nasir Uddin} and Odelia Schwartz",
year = "2019",
month = "4",
day = "1",
doi = "10.1016/j.conb.2019.01.008",
language = "English (US)",
volume = "55",
pages = "65--72",
journal = "Current Opinion in Neurobiology",
issn = "0959-4388",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Normalization and pooling in hierarchical models of natural images

AU - Sanchez-Giraldo, Luis G.

AU - Laskar, Md Nasir Uddin

AU - Schwartz, Odelia

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Divisive normalization and subunit pooling are two canonical classes of computation that have become widely used in descriptive (what) models of visual cortical processing. Normative (why) models from natural image statistics can help constrain the form and parameters of such classes of models. We focus on recent advances in two particular directions, namely deriving richer forms of divisive normalization, and advances in learning pooling from image statistics. We discuss the incorporation of such components into hierarchical models. We consider both hierarchical unsupervised learning from image statistics, and discriminative supervised learning in deep convolutional neural networks (CNNs). We further discuss studies on the utility and extensions of the convolutional architecture, which has also been adopted by recent descriptive models. We review the recent literature and discuss the current promises and gaps of using such approaches to gain a better understanding of how cortical neurons represent and process complex visual stimuli.

AB - Divisive normalization and subunit pooling are two canonical classes of computation that have become widely used in descriptive (what) models of visual cortical processing. Normative (why) models from natural image statistics can help constrain the form and parameters of such classes of models. We focus on recent advances in two particular directions, namely deriving richer forms of divisive normalization, and advances in learning pooling from image statistics. We discuss the incorporation of such components into hierarchical models. We consider both hierarchical unsupervised learning from image statistics, and discriminative supervised learning in deep convolutional neural networks (CNNs). We further discuss studies on the utility and extensions of the convolutional architecture, which has also been adopted by recent descriptive models. We review the recent literature and discuss the current promises and gaps of using such approaches to gain a better understanding of how cortical neurons represent and process complex visual stimuli.

UR - http://www.scopus.com/inward/record.url?scp=85061620186&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061620186&partnerID=8YFLogxK

U2 - 10.1016/j.conb.2019.01.008

DO - 10.1016/j.conb.2019.01.008

M3 - Review article

C2 - 30785005

AN - SCOPUS:85061620186

VL - 55

SP - 65

EP - 72

JO - Current Opinion in Neurobiology

JF - Current Opinion in Neurobiology

SN - 0959-4388

ER -