Deep minimax probability machine

Lirong He, Ziyi Guo, Kaizhu Huang, Zenglin Xu

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

2 Citations (Scopus)

Abstract

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.

Original languageEnglish
Title of host publicationProceedings - 19th IEEE International Conference on Data Mining Workshops, ICDMW 2019
EditorsPanagiotis Papapetrou, Xueqi Cheng, Qing He
PublisherIEEE Computer Society
Pages869-876
Number of pages8
ISBN (Electronic)9781728146034
DOIs
Publication statusPublished - Nov 2019
Event19th IEEE International Conference on Data Mining Workshops, ICDMW 2019 - Beijing, China
Duration: 8 Nov 201911 Nov 2019

Publication series

NameIEEE International Conference on Data Mining Workshops, ICDMW
Volume2019-November
ISSN (Print)2375-9232
ISSN (Electronic)2375-9259

Conference

Conference19th IEEE International Conference on Data Mining Workshops, ICDMW 2019
Country/TerritoryChina
CityBeijing
Period8/11/1911/11/19

Keywords

  • Adversarial attacks
  • Deep neural networks
  • Mimimax probability machine

Fingerprint

Dive into the research topics of 'Deep minimax probability machine'. Together they form a unique fingerprint.

Cite this