[4G4-GS-2m] 機械学習：学習方略(2/2)

[4G4-GS-2m-03] On the Effectiveness of Adversarial Examples for One-Class Classifier based on GAN

〇Yoshihiro Koseki¹ (1. Mitsubishi Electric)

Keywords:Adversarial Examples

Adversarial Examples are malicious input data created by adding a small perturbation to original input data, and these data make a classifier outputs a wrong result. This kind of malicious inputs are serious concerns for safety and security of AI systems being used in real world situations. In this work, we show an experimental result which indicates effectiveness and ineffectiveness of adversarial examples based on fast gradient sign method attack against a one-class classifier using reconstruction loss of generative adversarial network.