Learning, invariance, and generalization in high-order neural networks

Lee C. Giles, Tom Maxwell

Research output: Contribution to journalArticle

432 Citations (Scopus)

Abstract

High-order neural networks have been shown to have impressive computational, storage, and learning capabilities. This performance is because the order or structure of a high-order neural network can be tailored to the order or structure of a problem. Thus, a neural network designed for a particular class of problems becomes specialized but also very efficient in solving those problems. Furthermore, a priori knowledge, such as geometric invariances, can be encoded in high-order networks. Because this knowledge does not have to be learned, these networks are very efficient in solving problems that utilize this knowledge.

Original languageEnglish (US)
Pages (from-to)4972-4978
Number of pages7
JournalApplied Optics
Volume26
Issue number23
DOIs
StatePublished - Dec 1987

Fingerprint

Invariance
learning
invariance
Neural networks
problem solving

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics

Cite this

Giles, Lee C. ; Maxwell, Tom. / Learning, invariance, and generalization in high-order neural networks. In: Applied Optics. 1987 ; Vol. 26, No. 23. pp. 4972-4978.
@article{d02f6834c186499e8f13a416932a9d9a,
title = "Learning, invariance, and generalization in high-order neural networks",
abstract = "High-order neural networks have been shown to have impressive computational, storage, and learning capabilities. This performance is because the order or structure of a high-order neural network can be tailored to the order or structure of a problem. Thus, a neural network designed for a particular class of problems becomes specialized but also very efficient in solving those problems. Furthermore, a priori knowledge, such as geometric invariances, can be encoded in high-order networks. Because this knowledge does not have to be learned, these networks are very efficient in solving problems that utilize this knowledge.",
author = "Giles, {Lee C.} and Tom Maxwell",
year = "1987",
month = "12",
doi = "10.1364/AO.26.004972",
language = "English (US)",
volume = "26",
pages = "4972--4978",
journal = "Applied Optics",
issn = "1559-128X",
publisher = "The Optical Society",
number = "23",

}

Learning, invariance, and generalization in high-order neural networks. / Giles, Lee C.; Maxwell, Tom.

In: Applied Optics, Vol. 26, No. 23, 12.1987, p. 4972-4978.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Learning, invariance, and generalization in high-order neural networks

AU - Giles, Lee C.

AU - Maxwell, Tom

PY - 1987/12

Y1 - 1987/12

N2 - High-order neural networks have been shown to have impressive computational, storage, and learning capabilities. This performance is because the order or structure of a high-order neural network can be tailored to the order or structure of a problem. Thus, a neural network designed for a particular class of problems becomes specialized but also very efficient in solving those problems. Furthermore, a priori knowledge, such as geometric invariances, can be encoded in high-order networks. Because this knowledge does not have to be learned, these networks are very efficient in solving problems that utilize this knowledge.

AB - High-order neural networks have been shown to have impressive computational, storage, and learning capabilities. This performance is because the order or structure of a high-order neural network can be tailored to the order or structure of a problem. Thus, a neural network designed for a particular class of problems becomes specialized but also very efficient in solving those problems. Furthermore, a priori knowledge, such as geometric invariances, can be encoded in high-order networks. Because this knowledge does not have to be learned, these networks are very efficient in solving problems that utilize this knowledge.

UR - http://www.scopus.com/inward/record.url?scp=0023513717&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0023513717&partnerID=8YFLogxK

U2 - 10.1364/AO.26.004972

DO - 10.1364/AO.26.004972

M3 - Article

AN - SCOPUS:0023513717

VL - 26

SP - 4972

EP - 4978

JO - Applied Optics

JF - Applied Optics

SN - 1559-128X

IS - 23

ER -