Implementation of Reduced Precision Integer Epigenetic Networks in Hardware

Andrew Walter; Simon Jonathan Bale; Andy Tyrrell

Implementation of Reduced Precision Integer Epigenetic Networks in Hardware

Andrew Walter, Simon Jonathan Bale, Andy Tyrrell

Electronic Engineering

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

Abstract

This paper details the development of a resource efficient implementation of the Artificial Epigenetic Network (AEN) concept, based on reduced precision integer mathematics, and the translation of this implementation into hardware via a Field Programmable Gate Array (FPGA) to provide improvements in resource utilisation and execution speed while not sacrificing the unique benefits provided by the epigenetic mechanism. Validation of the implementation’s performance on the inverted pendulum task is obtained and compared to that of previous AENs, as well as experiments to determine how far the precision of the network may be reduced while still maintaining an acceptable degree of performance.

Original language	English
Title of host publication	IEEE International Conference on Evolvable Systems
Place of Publication	Orlando
Publisher	IEEE
Publication status	Published - 6 Dec 2021

Keywords

Artificial Epigenetic Networks
Field Programable Gate Array
Reduced Precision Computation
Digital Hardware

Access to Document

Implementation of Reduced Precision Integer Epigenetic Networks in Hardware (1)Final published version, 1.62 MB

Cite this

@inproceedings{c799cccb35694d1986a776a3f361e250,

title = "Implementation of Reduced Precision Integer Epigenetic Networks in Hardware",

abstract = "This paper details the development of a resource efficient implementation of the Artificial Epigenetic Network (AEN) concept, based on reduced precision integer mathematics, and the translation of this implementation into hardware via a Field Programmable Gate Array (FPGA) to provide improvements in resource utilisation and execution speed while not sacrificing the unique benefits provided by the epigenetic mechanism. Validation of the implementation{\textquoteright}s performance on the inverted pendulum task is obtained and compared to that of previous AENs, as well as experiments to determine how far the precision of the network may be reduced while still maintaining an acceptable degree of performance.",

keywords = "Artificial Epigenetic Networks, Field Programable Gate Array, Reduced Precision Computation, Digital Hardware",

author = "Andrew Walter and Bale, {Simon Jonathan} and Andy Tyrrell",

year = "2021",

month = dec,

day = "6",

language = "English",

booktitle = "IEEE International Conference on Evolvable Systems",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Implementation of Reduced Precision Integer Epigenetic Networks in Hardware

AU - Walter, Andrew

AU - Bale, Simon Jonathan

AU - Tyrrell, Andy

PY - 2021/12/6

Y1 - 2021/12/6

N2 - This paper details the development of a resource efficient implementation of the Artificial Epigenetic Network (AEN) concept, based on reduced precision integer mathematics, and the translation of this implementation into hardware via a Field Programmable Gate Array (FPGA) to provide improvements in resource utilisation and execution speed while not sacrificing the unique benefits provided by the epigenetic mechanism. Validation of the implementation’s performance on the inverted pendulum task is obtained and compared to that of previous AENs, as well as experiments to determine how far the precision of the network may be reduced while still maintaining an acceptable degree of performance.

AB - This paper details the development of a resource efficient implementation of the Artificial Epigenetic Network (AEN) concept, based on reduced precision integer mathematics, and the translation of this implementation into hardware via a Field Programmable Gate Array (FPGA) to provide improvements in resource utilisation and execution speed while not sacrificing the unique benefits provided by the epigenetic mechanism. Validation of the implementation’s performance on the inverted pendulum task is obtained and compared to that of previous AENs, as well as experiments to determine how far the precision of the network may be reduced while still maintaining an acceptable degree of performance.

KW - Artificial Epigenetic Networks

KW - Field Programable Gate Array

KW - Reduced Precision Computation

KW - Digital Hardware

M3 - Conference contribution

BT - IEEE International Conference on Evolvable Systems

PB - IEEE

CY - Orlando

ER -