Topological Evolution of Spiking Neural Networks

Samuel Slade; Li Zhang

doi:10.1109/ijcnn.2018.8489375

Topological Evolution of Spiking Neural Networks

Samuel Slade, Li Zhang

Computer and Information Sciences Department

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

3 Citations (Scopus)

27 Downloads (Pure)

Abstract

Neuro-evolution is often used to generate the parameters, topology, and rules of artificial neural networks. This technique allows for automatic configuration of a neural network. In this paper we propose a method to generate Spiking Neural Networks (SNNs) automatically called NENG (Neuro-Evolutionary Network Generation). The aim was to help alleviate the manual construction and optimization of neural network implementations. The results show the algorithm is successful at generating and improving the design of SNNs for a Classification task. After 812 generations with a population size of 20 the algorithm converges to model the Xor gate with 100% accuracy. The results show improvements to the algorithm execution time and number of neurons over time.

Original language	English
Title of host publication	IEEE World Congress on Computational Intelligence
Place of Publication	Piscataway
Publisher	IEEE
Pages	3135-3143
ISBN (Electronic)	9781509060146
ISBN (Print)	9781509060153
DOIs	https://doi.org/10.1109/ijcnn.2018.8489375
Publication status	Published - 8 Jul 2018
Event	IEEE World Congress on Computational Intelligence 2018 - Windsor Barra Convention Centre, Rio de Janeiro, Brazil Duration: 8 Jul 2018 → 13 Jul 2018

Conference

Conference	IEEE World Congress on Computational Intelligence 2018
Abbreviated title	WCCI 2018
Country/Territory	Brazil
City	Rio de Janeiro
Period	8/07/18 → 13/07/18

Access to Document

10.1109/ijcnn.2018.8489375

PID5334541Accepted author manuscript, 542 KB

Cite this

@inproceedings{0b8314f784e44e00aa97c6946c753eea,

title = "Topological Evolution of Spiking Neural Networks",

abstract = "Neuro-evolution is often used to generate the parameters, topology, and rules of artificial neural networks. This technique allows for automatic configuration of a neural network. In this paper we propose a method to generate Spiking Neural Networks (SNNs) automatically called NENG (Neuro-Evolutionary Network Generation). The aim was to help alleviate the manual construction and optimization of neural network implementations. The results show the algorithm is successful at generating and improving the design of SNNs for a Classification task. After 812 generations with a population size of 20 the algorithm converges to model the Xor gate with 100% accuracy. The results show improvements to the algorithm execution time and number of neurons over time.",

author = "Samuel Slade and Li Zhang",

year = "2018",

month = jul,

day = "8",

doi = "10.1109/ijcnn.2018.8489375",

language = "English",

isbn = "9781509060153",

pages = "3135--3143",

booktitle = "IEEE World Congress on Computational Intelligence",

publisher = "IEEE",

address = "United States",

note = "IEEE World Congress on Computational Intelligence 2018, WCCI 2018 ; Conference date: 08-07-2018 Through 13-07-2018",

}

TY - GEN

T1 - Topological Evolution of Spiking Neural Networks

AU - Slade, Samuel

AU - Zhang, Li

PY - 2018/7/8

Y1 - 2018/7/8

N2 - Neuro-evolution is often used to generate the parameters, topology, and rules of artificial neural networks. This technique allows for automatic configuration of a neural network. In this paper we propose a method to generate Spiking Neural Networks (SNNs) automatically called NENG (Neuro-Evolutionary Network Generation). The aim was to help alleviate the manual construction and optimization of neural network implementations. The results show the algorithm is successful at generating and improving the design of SNNs for a Classification task. After 812 generations with a population size of 20 the algorithm converges to model the Xor gate with 100% accuracy. The results show improvements to the algorithm execution time and number of neurons over time.

AB - Neuro-evolution is often used to generate the parameters, topology, and rules of artificial neural networks. This technique allows for automatic configuration of a neural network. In this paper we propose a method to generate Spiking Neural Networks (SNNs) automatically called NENG (Neuro-Evolutionary Network Generation). The aim was to help alleviate the manual construction and optimization of neural network implementations. The results show the algorithm is successful at generating and improving the design of SNNs for a Classification task. After 812 generations with a population size of 20 the algorithm converges to model the Xor gate with 100% accuracy. The results show improvements to the algorithm execution time and number of neurons over time.

U2 - 10.1109/ijcnn.2018.8489375

DO - 10.1109/ijcnn.2018.8489375

M3 - Conference contribution

SN - 9781509060153

SP - 3135

EP - 3143

BT - IEEE World Congress on Computational Intelligence

PB - IEEE

CY - Piscataway

T2 - IEEE World Congress on Computational Intelligence 2018

Y2 - 8 July 2018 through 13 July 2018

ER -

Topological Evolution of Spiking Neural Networks

Abstract

Conference

Access to Document

Fingerprint

Cite this