Abstract | ||
---|---|---|
This paper presents a hardware based implementation of a biologically-faithful astrocyte-based selfrepairing mechanism for Spiking Neural Networks. Spiking Astrocyte-neuron Networks (SANNs) are a new computing paradigm which capture the key mechanisms of how the human brain performs repairs. Using SANN in hardware affords the potential for realizing computing architecture that can self-repair. This paper demonstrates that Spiking Astrocyte Neural Network (SANN) in hardware have a resilience to significant levels of faults. The key novelty of the paper resides in implementing an SANN on FPGAs using fixed-point representation and demonstrating graceful performance degradation to different levels of injected faults via its self-repair capability. A fixed-point implementation of astrocyte, neurons and tripartite synapses are presented and compared against previous hardware floating-point and Matlab software implementations of SANN. All results are obtained from the SANN FPGA implementation and show how the reduced fixedpoint representation can maintain the biologically-realistic repair capability. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/ISVLSI.2017.80 | 2017 IEEE Computer Society Annual Symposium on VLSI (ISVLSI) |
Keywords | Field | DocType |
Self-repair,Astrocytes,Spiking neural networks,FPGA,Bio-inspired computing | MATLAB,Computer science,Bio-inspired computing,Field-programmable gate array,Novelty,Artificial neural network,Spiking neural network,Maintenance engineering,Embedded system,Self repair | Conference |
ISBN | Citations | PageRank |
978-1-5090-6763-3 | 0 | 0.34 |
References | Authors | |
13 | 10 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shvan Karim | 1 | 8 | 1.80 |
Jim Harkin | 2 | 325 | 36.82 |
Liam Mcdaid | 3 | 270 | 30.48 |
Bryan Gardiner | 4 | 28 | 8.31 |
Junxiu Liu | 5 | 125 | 23.91 |
David M. Halliday | 6 | 95 | 15.07 |
Andy M. Tyrrell | 7 | 629 | 73.61 |
Jon Timmis | 8 | 1237 | 120.32 |
Alan G. Millard | 9 | 24 | 5.94 |
Anju P. Johnson | 10 | 39 | 5.20 |