國立虎尾科技大學 |

Parameterizing and Aggregating Activation Functions in Deep Neural Networks.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Parameterizing and Aggregating Activation Functions in Deep Neural Networks./
Author:	Godfrey, Luke B.
Description:	1 online resource (126 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9780355842098

Parameterizing and Aggregating Activation Functions in Deep Neural Networks.
Godfrey, Luke B.

Parameterizing and Aggregating Activation Functions in Deep Neural Networks. - 1 online resource (126 pages)

Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.

Thesis (Ph.D.)--University of Arkansas, 2018.

Includes bibliographical references

The nonlinear activation functions applied by each neuron in a neural network are essential for making neural networks powerful representational models. If these are omitted, even deep neural networks reduce to simple linear regression due to the fact that a linear combi- nation of linear combinations is still a linear combination. In much of the existing literature on neural networks, just one or two activation functions are selected for the entire network, even though the use of heterogenous activation functions has been shown to produce su- perior results in some cases. Even less often employed are activation functions that can adapt their nonlinearities as network parameters along with standard weights and biases. This dissertation presents a collection of papers that advance the state of heterogenous and parameterized activation functions. Contributions of this dissertation include.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355842098Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Parameterizing and Aggregating Activation Functions in Deep Neural Networks.
LDR:02520ntm a2200373K 4500 001 915366
005 20180727125214.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355842098
035 $a (MiAaPQ)AAI10788695
035 $a (MiAaPQ)uark:12902
035 $a AAI10788695
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Godfrey, Luke B. $3 1188694
245 1 0 $a Parameterizing and Aggregating Activation Functions in Deep Neural Networks.
264 0 $c 2018
300 $a 1 online resource (126 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.
500 $a Adviser: Michael S. Gashler.
502 $a Thesis (Ph.D.)--University of Arkansas, 2018.
504 $a Includes bibliographical references
520 $a The nonlinear activation functions applied by each neuron in a neural network are essential for making neural networks powerful representational models. If these are omitted, even deep neural networks reduce to simple linear regression due to the fact that a linear combi- nation of linear combinations is still a linear combination. In much of the existing literature on neural networks, just one or two activation functions are selected for the entire network, even though the use of heterogenous activation functions has been shown to produce su- perior results in some cases. Even less often employed are activation functions that can adapt their nonlinearities as network parameters along with standard weights and biases. This dissertation presents a collection of papers that advance the state of heterogenous and parameterized activation functions. Contributions of this dissertation include.
520 $a • three novel parametric activation functions and applications of each,
520 $a • a study evaluating the utility of the parameters in parametric activation functions,
520 $a • an aggregated activation approach to modeling time-series data as an alternative to recurrent neural networks, and.
520 $a • an improvement upon existing work that aggregates neuron inputs using product in- stead of sum.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Arkansas. $b Computer Science. $3 1188695
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10788695 $z click for full text (PQDT)