Evasion attack of multi-class linear classifiers

Authors:
Han Xiao;Thomas Stibor;Claudia Eckert
Affiliations:
CeDoSIA of TUM Graduate School, Technische Universität München, Germany,IT Security, Technische Universität München, Germany;IT Security, Technische Universität München, Germany;IT Security, Technische Universität München, Germany
Venue:
PAKDD'12 Proceedings of the 16th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining - Volume Part I
Year:
2012

Citing 9
Cited 0

Applications of Data Mining in Computer Security

Applications of Data Mining in Computer Security
On the Learnability and Design of Output Codes for Multiclass Problems

Machine Learning
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
Machine Learning and Data Mining for Computer Security: Methods and Applications (Advanced Information and Knowledge Processing)

Machine Learning and Data Mining for Computer Security: Methods and Applications (Advanced Information and Knowledge Processing)
Spam Filtering Using Statistical Data Compression Models

The Journal of Machine Learning Research
Spam Filtering Based On The Analysis Of Text Information Embedded Into Images

The Journal of Machine Learning Research
A sequential dual method for large scale multi-class linear svms

Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining
LIBLINEAR: A Library for Large Linear Classification

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

Machine learning has yield significant advances in decision-making for complex systems, but are they robust against adversarial attacks? We generalize the evasion attack problem to the multi-class linear classifiers, and present an efficient algorithm for approximating the optimal disguised instance. Experiments on real-world data demonstrate the effectiveness of our method.