TY - GEN
T1 - Deep minimax probability machine
AU - He, Lirong
AU - Guo, Ziyi
AU - Huang, Kaizhu
AU - Xu, Zenglin
N1 - Funding Information:
This paper was in part supported by Grants from the Natural Science Foundation of China(No. 61572111), and two Fundamental Research Funds for the Central Universities of China (Nos.ZYGX2016Z003, ZYGX2017KYQD177),
Publisher Copyright:
© 2019 IEEE.
PY - 2019/11
Y1 - 2019/11
N2 - Deep neural networks enjoy a powerful representation and have proven effective in a number of applications. However, recent advances show that deep neural networks are vulnerable to adversarial attacks incurred by the so-called adversarial examples. Although the adversarial example is only slightly different from the input sample, the neural network classifies it as the wrong class. In order to alleviate this problem, we propose the Deep Minimax Probability Machine (DeepMPM), which applies MPM to deep neural networks in an end-to-end fashion. In a worst-case scenario, MPM tries to minimize an upper bound of misclassification probabilities, considering the global information (i.e., mean and covariance information of each class). DeepMPM can be more robust since it learns the worst-case bound on the probability of misclassification of future data.
AB - Deep neural networks enjoy a powerful representation and have proven effective in a number of applications. However, recent advances show that deep neural networks are vulnerable to adversarial attacks incurred by the so-called adversarial examples. Although the adversarial example is only slightly different from the input sample, the neural network classifies it as the wrong class. In order to alleviate this problem, we propose the Deep Minimax Probability Machine (DeepMPM), which applies MPM to deep neural networks in an end-to-end fashion. In a worst-case scenario, MPM tries to minimize an upper bound of misclassification probabilities, considering the global information (i.e., mean and covariance information of each class). DeepMPM can be more robust since it learns the worst-case bound on the probability of misclassification of future data.
KW - Adversarial attacks
KW - Deep neural networks
KW - Mimimax probability machine
UR - http://www.scopus.com/inward/record.url?scp=85078764095&partnerID=8YFLogxK
U2 - 10.1109/ICDMW.2019.00127
DO - 10.1109/ICDMW.2019.00127
M3 - Conference Proceeding
AN - SCOPUS:85078764095
T3 - IEEE International Conference on Data Mining Workshops, ICDMW
SP - 869
EP - 876
BT - Proceedings - 19th IEEE International Conference on Data Mining Workshops, ICDMW 2019
A2 - Papapetrou, Panagiotis
A2 - Cheng, Xueqi
A2 - He, Qing
PB - IEEE Computer Society
T2 - 19th IEEE International Conference on Data Mining Workshops, ICDMW 2019
Y2 - 8 November 2019 through 11 November 2019
ER -