Direct Fit to Nature: An Evolutionary Perspective on Biological and Artificial Neural Networks

Uri Hasson; Samuel A. Nastase; Ariel Goldstein

doi:10.1016/j.neuron.2019.12.002

Direct Fit to Nature: An Evolutionary Perspective on Biological and Artificial Neural Networks

Uri Hasson, Samuel A. Nastase, Ariel Goldstein

Research output: Contribution to journal › Review article › peer-review

254 Scopus citations

Abstract

To guide adaptive behavior and support predictions in real-life contexts, the brain may rely on opaque, over-parameterized models capable of directly fitting to the multidimensional world, while being blind—like evolution—to the underlying rules and causes.

Original language	English (US)
Pages (from-to)	416-434
Number of pages	19
Journal	Neuron
Volume	105
Issue number	3
DOIs	https://doi.org/10.1016/j.neuron.2019.12.002
State	Published - Feb 5 2020

All Science Journal Classification (ASJC) codes

General Neuroscience

Keywords

evolution
experimental design
interpolation
learning
neural networks

Access to Document

10.1016/j.neuron.2019.12.002

Cite this

@article{28e69e03374642de89a08a7124408827,

title = "Direct Fit to Nature: An Evolutionary Perspective on Biological and Artificial Neural Networks",

abstract = "To guide adaptive behavior and support predictions in real-life contexts, the brain may rely on opaque, over-parameterized models capable of directly fitting to the multidimensional world, while being blind—like evolution—to the underlying rules and causes.",

keywords = "evolution, experimental design, interpolation, learning, neural networks",

author = "Uri Hasson and Nastase, \{Samuel A.\} and Ariel Goldstein",

note = "Publisher Copyright: {\textcopyright} 2019 Elsevier Inc.",

year = "2020",

month = feb,

day = "5",

doi = "10.1016/j.neuron.2019.12.002",

language = "English (US)",

volume = "105",

pages = "416--434",

journal = "Neuron",

issn = "0896-6273",

publisher = "Cell Press",

number = "3",

}

TY - JOUR

T1 - Direct Fit to Nature

T2 - An Evolutionary Perspective on Biological and Artificial Neural Networks

AU - Hasson, Uri

AU - Nastase, Samuel A.

AU - Goldstein, Ariel

PY - 2020/2/5

Y1 - 2020/2/5

N2 - To guide adaptive behavior and support predictions in real-life contexts, the brain may rely on opaque, over-parameterized models capable of directly fitting to the multidimensional world, while being blind—like evolution—to the underlying rules and causes.

AB - To guide adaptive behavior and support predictions in real-life contexts, the brain may rely on opaque, over-parameterized models capable of directly fitting to the multidimensional world, while being blind—like evolution—to the underlying rules and causes.

KW - evolution

KW - experimental design

KW - interpolation

KW - learning

KW - neural networks

UR - https://www.scopus.com/pages/publications/85078699246

UR - https://www.scopus.com/inward/citedby.url?scp=85078699246&partnerID=8YFLogxK

U2 - 10.1016/j.neuron.2019.12.002

DO - 10.1016/j.neuron.2019.12.002

M3 - Review article

C2 - 32027833

AN - SCOPUS:85078699246

SN - 0896-6273

VL - 105

SP - 416

EP - 434

JO - Neuron

JF - Neuron

IS - 3

ER -

Direct Fit to Nature: An Evolutionary Perspective on Biological and Artificial Neural Networks

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this