Classifier evasion: models and open problems

Authors:
Blaine Nelson;Benjamin I. P. Rubinstein;Ling Huang;Anthony D. Joseph;J. D. Tygar
Affiliations:
UC Berkeley;Microsoft Research;Intel Labs Berkeley;UC Berkeley and Intel Labs Berkeley;UC Berkeley
Venue:
PSDML'10 Proceedings of the international ECML/PKDD conference on Privacy and security issues in data mining and machine learning
Year:
2010

Citing 6
Cited 3

Solving convex programs by random walks

Journal of the ACM (JACM)
Adversarial classification

Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
Adversarial learning

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Prediction, Learning, and Games

Prediction, Learning, and Games
The security of machine learning

Machine Learning
Multiple classifier systems under attack

MCS'10 Proceedings of the 9th international conference on Multiple Classifier Systems

Adversarial machine learning

Proceedings of the 4th ACM workshop on Security and artificial intelligence
Query strategies for evading convex-inducing classifiers

The Journal of Machine Learning Research
Early security classification of skype users via machine learning

Proceedings of the 2013 ACM workshop on Artificial intelligence and security

Quantified Score

Hi-index	0.00

Visualization

Abstract

As a growing number of software developers apply machine learning to make key decisions in their systems, adversaries are adapting and launching ever more sophisticated attacks against these systems. The near-optimal evasion problem considers an adversary that searches for a low-cost negative instance by submitting a minimal number of queries to a classifier, in order to effectively evade the classifier. In this position paper, we posit several open problems and alternative variants to the near-optimal evasion problem. Solutions to these problems would significantly advance the state-of-the-art in secure machine learning.