Adversarial Training for Probabilistic Spiking Neural Networks

TY - CHAP

T1 - Adversarial Training for Probabilistic Spiking Neural Networks

AU - Bagheri, Alireza

AU - Simeone, Osvaldo

AU - Rajendran, Bipin

PY - 2018/8/24

Y1 - 2018/8/24

N2 - Classifiers trained using conventional empirical risk minimization or maximum likelihood methods are known to suffer dramatic performance degradations when tested over examples adversarially selected based on knowledge of the clas-sifier's decision rule. Due to the prominence of Artificial Neural Networks (ANNs) as classifiers, their sensitivity to adversarial examples, as well as robust training schemes, have been recently the subject of intense investigation. In this paper, for the first time, the sensitivity of spiking neural networks (SNNs), or third-generation neural networks, to adversarial examples is studied. The study considers rate and time encoding, as well as rate and first-to-spike decoding. Furthermore, a robust training mechanism is proposed that is demonstrated to enhance the performance of SNNs under white-box attacks.

AB - Classifiers trained using conventional empirical risk minimization or maximum likelihood methods are known to suffer dramatic performance degradations when tested over examples adversarially selected based on knowledge of the clas-sifier's decision rule. Due to the prominence of Artificial Neural Networks (ANNs) as classifiers, their sensitivity to adversarial examples, as well as robust training schemes, have been recently the subject of intense investigation. In this paper, for the first time, the sensitivity of spiking neural networks (SNNs), or third-generation neural networks, to adversarial examples is studied. The study considers rate and time encoding, as well as rate and first-to-spike decoding. Furthermore, a robust training mechanism is proposed that is demonstrated to enhance the performance of SNNs under white-box attacks.

KW - adversarial examples

KW - adversarial training

KW - Generalized Linear Model (GLM)

KW - Spiking Neural Networks (SNNs)

UR - http://www.scopus.com/inward/record.url?scp=85053456439&partnerID=8YFLogxK

U2 - 10.1109/SPAWC.2018.8446003

DO - 10.1109/SPAWC.2018.8446003

M3 - Conference paper

AN - SCOPUS:85053456439

SN - 9781538635124

VL - 2018-June

BT - 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 19th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2018

Y2 - 25 June 2018 through 28 June 2018

ER -