Catherine Schuman, J. Parker Mitchell, J. Travis Johnston, Maryam Parsa, Bill Kay, Prasanna Date and Robert M. Patton

July, 2020

IJCNN: The International Joint Conference on Neural Networks

https://www.ijcnn.org/

View Article

Abstract

Though robustness and resilience are commonly quoted as features of neuromorphic computing systems, the expected performance of neuromorphic systems in the face of hardware failures is not clear. In this work, we study the effect of failures on the performance of four different training algorithms for spiking neural networks on neuromorphic systems: two back-propagation-based training approaches (Whetstone and SLAYER), a liquid state machine or reservoir computing approach, and an evolutionary optimization-based approach (EONS). We show that these four different approaches have very different resilience characteristics with respect to simulated hardware failures. We then analyze an approach for training more resilient spiking neural networks using the evolutionary optimization approach. We show how this approach produces more resilient networks and discuss how it can be extended to other spiking neural network training approaches as well.

Citation Information

Text

author      C. D. Schuman and J. P. Mitchell and J. T. Johnston and M. Parsa and B. Kay
            and P. Date and R. M. Patton
title       Resilience and Robustness of Spiking Neural Networks for Neuromorphic Systems
booktitle   IJCNN: The International Joint Conference on Neural Networks
year        2020

Bibtex

@INPROCEEDINGS{smj:20:rrs,
    author = "C. D. Schuman and J. P. Mitchell and J. T. Johnston and M. Parsa and B. Kay
                and P. Date and R. M. Patton",
    title = "Resilience and Robustness of Spiking Neural Networks for Neuromorphic Systems",
    booktitle = "IJCNN: The International Joint Conference on Neural Networks",
    year = "2020"
}