Exploring Self-Repair in a Coupled Spiking Astrocyte Neural Network

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

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

It is now known that astrocytes modulate the activity at the tripartite synapses where indirect signaling via the retrograde messengers, endocannabinoids, leads to a localized self-repairing capability. In this paper, a self-repairing spiking astrocyte neural network (SANN) is proposed to demonstrate a distributed self-repairing capability at the network level. The SANN uses a novel learning rule that combines the spike-timing-dependent plasticity (STDP) and Bienenstock, Cooper, and Munro (BCM) learning rules (hereafter referred to as the BSTDP rule). In this learning rule, the synaptic weight potentiation is not only driven by the temporal difference between the presynaptic and postsynaptic neuron firing times but also by the postsynaptic neuron activity. We will show in this paper that the BSTDP modulates the height of the plasticity window to establish an input-output mapping (in the learning phase) and also maintains this mapping (via self-repair) if synaptic pathways become dysfunctional. It is the functional dependence of postsynaptic neuron firing activity on the height of the plasticity window that underpins how the proposed SANN self-repairs on the fly. The SANN also uses the coupling between the tripartite synapses and ɣ-GABAergic interneurons. This interaction gives rise to a presynaptic neuron frequency filtering capability that serves to route information, represented as spike trains, to different neurons in the subsequent layers of the SANN. The proposed SANN follows a feedforward architecture with multiple interneuron pathways and astrocytes modulate synaptic activity at the hidden and output neuronal layers. The self-repairing capability will be demonstrated in a robotic obstacle avoidance application, and the simulation results will show that the SANN can maintain learned maneuvers at synaptic fault densities of up to 80% regardless of the fault locations.
LanguageEnglish
Article number8423789
Pages865-875
Number of pages11
JournalIEEE Transactions on Neural Networks and Learning Systems
Volume30
Issue number3
Early online date31 Jul 2018
DOIs
Publication statusPublished - 1 Mar 2019
Externally publishedYes

Fingerprint

Repair
Neural networks
Neurons
Plasticity
Astrocytes
Electric fault location
Collision avoidance
Robotics

Cite this

Liu, Junxiu ; McDaid, Liam ; Harkin, Jim ; Karim, Shvan ; Johnson, Anju ; Millard, Alan ; Hilder, James ; Halliday, David ; Tyrrell, Andy ; Timmis, Jon. / Exploring Self-Repair in a Coupled Spiking Astrocyte Neural Network. In: IEEE Transactions on Neural Networks and Learning Systems. 2019 ; Vol. 30, No. 3. pp. 865-875.
@article{30761e4dc49b494caa527b377cad1d39,
title = "Exploring Self-Repair in a Coupled Spiking Astrocyte Neural Network",
abstract = "It is now known that astrocytes modulate the activity at the tripartite synapses where indirect signaling via the retrograde messengers, endocannabinoids, leads to a localized self-repairing capability. In this paper, a self-repairing spiking astrocyte neural network (SANN) is proposed to demonstrate a distributed self-repairing capability at the network level. The SANN uses a novel learning rule that combines the spike-timing-dependent plasticity (STDP) and Bienenstock, Cooper, and Munro (BCM) learning rules (hereafter referred to as the BSTDP rule). In this learning rule, the synaptic weight potentiation is not only driven by the temporal difference between the presynaptic and postsynaptic neuron firing times but also by the postsynaptic neuron activity. We will show in this paper that the BSTDP modulates the height of the plasticity window to establish an input-output mapping (in the learning phase) and also maintains this mapping (via self-repair) if synaptic pathways become dysfunctional. It is the functional dependence of postsynaptic neuron firing activity on the height of the plasticity window that underpins how the proposed SANN self-repairs on the fly. The SANN also uses the coupling between the tripartite synapses and ɣ-GABAergic interneurons. This interaction gives rise to a presynaptic neuron frequency filtering capability that serves to route information, represented as spike trains, to different neurons in the subsequent layers of the SANN. The proposed SANN follows a feedforward architecture with multiple interneuron pathways and astrocytes modulate synaptic activity at the hidden and output neuronal layers. The self-repairing capability will be demonstrated in a robotic obstacle avoidance application, and the simulation results will show that the SANN can maintain learned maneuvers at synaptic fault densities of up to 80{\%} regardless of the fault locations.",
keywords = "Synapses, Neurons, Robots, Fault tolerance, Fault tolerant systems, Hardware, Biological neural networks, Astrocyte, Self-repair, Obstacle avoidance, Spiking neural networks, obstacle avoidance, fault tolerance, spiking neural networks, self-repair",
author = "Junxiu Liu and Liam McDaid and Jim Harkin and Shvan Karim and Anju Johnson and Alan Millard and James Hilder and David Halliday and Andy Tyrrell and Jon Timmis",
year = "2019",
month = "3",
day = "1",
doi = "10.1109/TNNLS.2018.2854291",
language = "English",
volume = "30",
pages = "865--875",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "3",

}

Liu, J, McDaid, L, Harkin, J, Karim, S, Johnson, A, Millard, A, Hilder, J, Halliday, D, Tyrrell, A & Timmis, J 2019, 'Exploring Self-Repair in a Coupled Spiking Astrocyte Neural Network', IEEE Transactions on Neural Networks and Learning Systems, vol. 30, no. 3, 8423789, pp. 865-875. https://doi.org/10.1109/TNNLS.2018.2854291

Exploring Self-Repair in a Coupled Spiking Astrocyte Neural Network. / Liu, Junxiu; McDaid, Liam; Harkin, Jim; Karim, Shvan; Johnson, Anju; Millard, Alan; Hilder, James; Halliday, David; Tyrrell, Andy ; Timmis, Jon.

In: IEEE Transactions on Neural Networks and Learning Systems, Vol. 30, No. 3, 8423789, 01.03.2019, p. 865-875.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Exploring Self-Repair in a Coupled Spiking Astrocyte Neural Network

AU - Liu, Junxiu

AU - McDaid, Liam

AU - Harkin, Jim

AU - Karim, Shvan

AU - Johnson, Anju

AU - Millard, Alan

AU - Hilder, James

AU - Halliday, David

AU - Tyrrell, Andy

AU - Timmis, Jon

PY - 2019/3/1

Y1 - 2019/3/1

N2 - It is now known that astrocytes modulate the activity at the tripartite synapses where indirect signaling via the retrograde messengers, endocannabinoids, leads to a localized self-repairing capability. In this paper, a self-repairing spiking astrocyte neural network (SANN) is proposed to demonstrate a distributed self-repairing capability at the network level. The SANN uses a novel learning rule that combines the spike-timing-dependent plasticity (STDP) and Bienenstock, Cooper, and Munro (BCM) learning rules (hereafter referred to as the BSTDP rule). In this learning rule, the synaptic weight potentiation is not only driven by the temporal difference between the presynaptic and postsynaptic neuron firing times but also by the postsynaptic neuron activity. We will show in this paper that the BSTDP modulates the height of the plasticity window to establish an input-output mapping (in the learning phase) and also maintains this mapping (via self-repair) if synaptic pathways become dysfunctional. It is the functional dependence of postsynaptic neuron firing activity on the height of the plasticity window that underpins how the proposed SANN self-repairs on the fly. The SANN also uses the coupling between the tripartite synapses and ɣ-GABAergic interneurons. This interaction gives rise to a presynaptic neuron frequency filtering capability that serves to route information, represented as spike trains, to different neurons in the subsequent layers of the SANN. The proposed SANN follows a feedforward architecture with multiple interneuron pathways and astrocytes modulate synaptic activity at the hidden and output neuronal layers. The self-repairing capability will be demonstrated in a robotic obstacle avoidance application, and the simulation results will show that the SANN can maintain learned maneuvers at synaptic fault densities of up to 80% regardless of the fault locations.

AB - It is now known that astrocytes modulate the activity at the tripartite synapses where indirect signaling via the retrograde messengers, endocannabinoids, leads to a localized self-repairing capability. In this paper, a self-repairing spiking astrocyte neural network (SANN) is proposed to demonstrate a distributed self-repairing capability at the network level. The SANN uses a novel learning rule that combines the spike-timing-dependent plasticity (STDP) and Bienenstock, Cooper, and Munro (BCM) learning rules (hereafter referred to as the BSTDP rule). In this learning rule, the synaptic weight potentiation is not only driven by the temporal difference between the presynaptic and postsynaptic neuron firing times but also by the postsynaptic neuron activity. We will show in this paper that the BSTDP modulates the height of the plasticity window to establish an input-output mapping (in the learning phase) and also maintains this mapping (via self-repair) if synaptic pathways become dysfunctional. It is the functional dependence of postsynaptic neuron firing activity on the height of the plasticity window that underpins how the proposed SANN self-repairs on the fly. The SANN also uses the coupling between the tripartite synapses and ɣ-GABAergic interneurons. This interaction gives rise to a presynaptic neuron frequency filtering capability that serves to route information, represented as spike trains, to different neurons in the subsequent layers of the SANN. The proposed SANN follows a feedforward architecture with multiple interneuron pathways and astrocytes modulate synaptic activity at the hidden and output neuronal layers. The self-repairing capability will be demonstrated in a robotic obstacle avoidance application, and the simulation results will show that the SANN can maintain learned maneuvers at synaptic fault densities of up to 80% regardless of the fault locations.

KW - Synapses

KW - Neurons

KW - Robots

KW - Fault tolerance

KW - Fault tolerant systems

KW - Hardware

KW - Biological neural networks

KW - Astrocyte

KW - Self-repair

KW - Obstacle avoidance

KW - Spiking neural networks

KW - obstacle avoidance

KW - fault tolerance

KW - spiking neural networks

KW - self-repair

UR - http://www.scopus.com/inward/record.url?scp=85050974272&partnerID=8YFLogxK

U2 - 10.1109/TNNLS.2018.2854291

DO - 10.1109/TNNLS.2018.2854291

M3 - Article

VL - 30

SP - 865

EP - 875

JO - IEEE Transactions on Neural Networks and Learning Systems

T2 - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 3

M1 - 8423789

ER -