Mining adversarial patterns via regularized loss minimization |
| |
Authors: | Wei Liu Sanjay Chawla |
| |
Affiliation: | (1) Carnegie Mellon University, Pittsburgh, PA 15232, USA;(2) Knight Capital Group, Jersey city, NJ, USA;(3) IBM Research, York Town, NY, USA;(4) University of Southern California, Los Angeles, CA 90089, USA;(5) Bar-Ilan University, Ramat-Gan, 52900, Israel;(6) University of Maryland, College Park, MD 20742, USA |
| |
Abstract: | Traditional classification methods assume that the training and the test data arise from the same underlying distribution.
However, in several adversarial settings, the test set is deliberately constructed in order to increase the error rates of
the classifier. A prominent example is spam email where words are transformed to get around word based features embedded in
a spam filter. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|