3

SoK: What Have We Learned About Black-box Attacks Against Classifiers?
Abstract Dozens of papers have been written to study the vulnerability of machine learning classifiers to adversarial examples in black-box settings in which the adversary only has API access to the target classifier.
Suya Fnu, Anshuman Suri, Tingwei Zhang, Jingtao Hong, Yuan Tian, David Evans
PDF Code
SoK: What Have We Learned About Black-box Attacks Against Classifiers?