Homeostatic Fault Tolerance in Spiking Neural Networks: A Dynamic Hardware Perspective

Anju Johnson, Junxiu Liu, Alan Millard, Shvan Karim, Andy Tyrrell, Jim Harkin, Jon Timmis, David Halliday

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Fault tolerance is a remarkable feature of biological systems and their self-repair capability influence modern electronic systems. In this paper, we propose a novel plastic
neural network model, which establishes homeostasis in a spiking neural network. Combined with this plasticity and the inspiration from inhibitory interneurons, we develop a fault-resilient robotic controller implemented on an FPGA establishing obstacle avoidance task. We demonstrate the proposed methodology on a spiking neural network implemented on Xilinx Artix-7 FPGA. The system is able to maintain stable firing (tolerance ±10%) with a loss of up to 75% of the original synaptic inputs to a neuron. Our repair mechanism has minimal hardware overhead with a tuning circuit (repair unit) which consumes only three slices/neuron for implementing a threshold voltagebased homeostatic fault-tolerant unit. The overall architecture has a minimal impact on power consumption and, therefore, supports scalable implementations. This paper opens a novel way of implementing the behavior of natural fault tolerant system in hardware establishing homeostatic self-repair behavior.
LanguageEnglish
Pages687-699
Number of pages13
JournalIEEE Transactions on Biomedical Circuits and Systems
Volume65
Issue number2
Early online date28 Jul 2017
DOIs
Publication statusPublished - Feb 2018
Externally publishedYes

Fingerprint

Fault tolerance
Repair
Neural networks
Hardware
Neurons
Field programmable gate arrays (FPGA)
Biological systems
Collision avoidance
Plasticity
Robotics
Electric power utilization
Tuning
Controllers
Networks (circuits)

Cite this

Johnson, Anju ; Liu, Junxiu ; Millard, Alan ; Karim, Shvan ; Tyrrell, Andy ; Harkin, Jim ; Timmis, Jon ; Halliday, David. / Homeostatic Fault Tolerance in Spiking Neural Networks : A Dynamic Hardware Perspective. In: IEEE Transactions on Biomedical Circuits and Systems. 2018 ; Vol. 65, No. 2. pp. 687-699.
@article{3a09b7cc3f3747558465cb41cedb5513,
title = "Homeostatic Fault Tolerance in Spiking Neural Networks: A Dynamic Hardware Perspective",
abstract = "Fault tolerance is a remarkable feature of biological systems and their self-repair capability influence modern electronic systems. In this paper, we propose a novel plasticneural network model, which establishes homeostasis in a spiking neural network. Combined with this plasticity and the inspiration from inhibitory interneurons, we develop a fault-resilient robotic controller implemented on an FPGA establishing obstacle avoidance task. We demonstrate the proposed methodology on a spiking neural network implemented on Xilinx Artix-7 FPGA. The system is able to maintain stable firing (tolerance ±10{\%}) with a loss of up to 75{\%} of the original synaptic inputs to a neuron. Our repair mechanism has minimal hardware overhead with a tuning circuit (repair unit) which consumes only three slices/neuron for implementing a threshold voltagebased homeostatic fault-tolerant unit. The overall architecture has a minimal impact on power consumption and, therefore, supports scalable implementations. This paper opens a novel way of implementing the behavior of natural fault tolerant system in hardware establishing homeostatic self-repair behavior.",
keywords = "Self-repair, Homeostasis, Fault tolerance, FPGA, Dynamic partial reconfiguration, Bio-inspired engineering, Mixed-mode clock manager, Phase locked loop",
author = "Anju Johnson and Junxiu Liu and Alan Millard and Shvan Karim and Andy Tyrrell and Jim Harkin and Jon Timmis and David Halliday",
year = "2018",
month = "2",
doi = "10.1109/TCSI.2017.2726763",
language = "English",
volume = "65",
pages = "687--699",
journal = "IEEE Transactions on Biomedical Circuits and Systems",
issn = "1932-4545",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

Homeostatic Fault Tolerance in Spiking Neural Networks : A Dynamic Hardware Perspective. / Johnson, Anju; Liu, Junxiu; Millard, Alan; Karim, Shvan; Tyrrell, Andy ; Harkin, Jim; Timmis, Jon; Halliday, David.

In: IEEE Transactions on Biomedical Circuits and Systems, Vol. 65, No. 2, 02.2018, p. 687-699.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Homeostatic Fault Tolerance in Spiking Neural Networks

T2 - IEEE Transactions on Biomedical Circuits and Systems

AU - Johnson, Anju

AU - Liu, Junxiu

AU - Millard, Alan

AU - Karim, Shvan

AU - Tyrrell, Andy

AU - Harkin, Jim

AU - Timmis, Jon

AU - Halliday, David

PY - 2018/2

Y1 - 2018/2

N2 - Fault tolerance is a remarkable feature of biological systems and their self-repair capability influence modern electronic systems. In this paper, we propose a novel plasticneural network model, which establishes homeostasis in a spiking neural network. Combined with this plasticity and the inspiration from inhibitory interneurons, we develop a fault-resilient robotic controller implemented on an FPGA establishing obstacle avoidance task. We demonstrate the proposed methodology on a spiking neural network implemented on Xilinx Artix-7 FPGA. The system is able to maintain stable firing (tolerance ±10%) with a loss of up to 75% of the original synaptic inputs to a neuron. Our repair mechanism has minimal hardware overhead with a tuning circuit (repair unit) which consumes only three slices/neuron for implementing a threshold voltagebased homeostatic fault-tolerant unit. The overall architecture has a minimal impact on power consumption and, therefore, supports scalable implementations. This paper opens a novel way of implementing the behavior of natural fault tolerant system in hardware establishing homeostatic self-repair behavior.

AB - Fault tolerance is a remarkable feature of biological systems and their self-repair capability influence modern electronic systems. In this paper, we propose a novel plasticneural network model, which establishes homeostasis in a spiking neural network. Combined with this plasticity and the inspiration from inhibitory interneurons, we develop a fault-resilient robotic controller implemented on an FPGA establishing obstacle avoidance task. We demonstrate the proposed methodology on a spiking neural network implemented on Xilinx Artix-7 FPGA. The system is able to maintain stable firing (tolerance ±10%) with a loss of up to 75% of the original synaptic inputs to a neuron. Our repair mechanism has minimal hardware overhead with a tuning circuit (repair unit) which consumes only three slices/neuron for implementing a threshold voltagebased homeostatic fault-tolerant unit. The overall architecture has a minimal impact on power consumption and, therefore, supports scalable implementations. This paper opens a novel way of implementing the behavior of natural fault tolerant system in hardware establishing homeostatic self-repair behavior.

KW - Self-repair

KW - Homeostasis

KW - Fault tolerance

KW - FPGA

KW - Dynamic partial reconfiguration

KW - Bio-inspired engineering

KW - Mixed-mode clock manager

KW - Phase locked loop

UR - http://www.scopus.com/inward/record.url?scp=85028947387&partnerID=8YFLogxK

U2 - 10.1109/TCSI.2017.2726763

DO - 10.1109/TCSI.2017.2726763

M3 - Article

VL - 65

SP - 687

EP - 699

JO - IEEE Transactions on Biomedical Circuits and Systems

JF - IEEE Transactions on Biomedical Circuits and Systems

SN - 1932-4545

IS - 2

ER -