Implementing Size-Optimal Discrete Neural Networks Requires Analog Circuitry

Valeriu Beiu

doi:European

Implementing Size-Optimal Discrete Neural Networks Requires Analog Circuitry

Valeriu Beiu

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper starts by overviewing results dealing with the approximation capabilities of neural networks, as well as bounds on the size of threshold gate circuits. Based on a constructive solution for Kolmogorov`s superpositions the authors show that implementing Boolean functions can be done using neurons having an identity transfer function. Because in this case the size of the network is minimized, it follows that size-optimal solutions for implementing Boolean functions can be obtained using analog circuitry. Conclusions and several comments on the required precision are ending the paper.

Original language	English
Title of host publication	European Signal Processing Conference
DOIs	https://doi.org/European
Publication status	Published - Sept 8 1998
Event	EuSiPCo'98 - Island of Rhodes, Greece Duration: Sept 8 1998 → …

Conference

Conference	EuSiPCo'98
Period	9/8/98 → …

Access to Document

European

http://www.osti.gov/scitech/biblio/291117

Cite this

@inproceedings{aeeb4daa9a1d419b99c3ae983efa4ea3,

title = "Implementing Size-Optimal Discrete Neural Networks Requires Analog Circuitry",

abstract = "This paper starts by overviewing results dealing with the approximation capabilities of neural networks, as well as bounds on the size of threshold gate circuits. Based on a constructive solution for Kolmogorov`s superpositions the authors show that implementing Boolean functions can be done using neurons having an identity transfer function. Because in this case the size of the network is minimized, it follows that size-optimal solutions for implementing Boolean functions can be obtained using analog circuitry. Conclusions and several comments on the required precision are ending the paper.",

author = "Valeriu Beiu",

year = "1998",

month = sep,

day = "8",

doi = "European",

language = "English",

booktitle = "European Signal Processing Conference",

note = "EuSiPCo'98 ; Conference date: 08-09-1998",

}

TY - GEN

T1 - Implementing Size-Optimal Discrete Neural Networks Requires Analog Circuitry

AU - Beiu, Valeriu

PY - 1998/9/8

Y1 - 1998/9/8

N2 - This paper starts by overviewing results dealing with the approximation capabilities of neural networks, as well as bounds on the size of threshold gate circuits. Based on a constructive solution for Kolmogorov`s superpositions the authors show that implementing Boolean functions can be done using neurons having an identity transfer function. Because in this case the size of the network is minimized, it follows that size-optimal solutions for implementing Boolean functions can be obtained using analog circuitry. Conclusions and several comments on the required precision are ending the paper.

AB - This paper starts by overviewing results dealing with the approximation capabilities of neural networks, as well as bounds on the size of threshold gate circuits. Based on a constructive solution for Kolmogorov`s superpositions the authors show that implementing Boolean functions can be done using neurons having an identity transfer function. Because in this case the size of the network is minimized, it follows that size-optimal solutions for implementing Boolean functions can be obtained using analog circuitry. Conclusions and several comments on the required precision are ending the paper.

U2 - European

DO - European

M3 - Conference contribution

BT - European Signal Processing Conference

T2 - EuSiPCo'98

Y2 - 8 September 1998

ER -

Implementing Size-Optimal Discrete Neural Networks Requires Analog Circuitry

Abstract

Conference

Access to Document

Fingerprint

Cite this