New and unseen polymorphic malware, zero-day attacks, or other types of advanced persistent threats are usually not detected by signature-based security devices, firewalls, or anti-viruses. This represents a challenge to the network security industry as the amount and variability of incidents has been increasing. Consequently, this complicates the design of learning-based detection systems relying on features extracted from network data. The problem is caused by different joint distribution of observation (features) and labels in the training and testing data sets. This paper proposes a classification system designed to detect both known as well as previouslyunseen security threats. The classifiers use statistical feature representation computed from the network traffic and learn to recognize malicious behavior. The representation is designed and optimized to be invariant to the most common changes of malware behaviors. This is achieved in part by a feature histogram constructed for each group of HTTP flows (proxy log records) of a user visiting a particular hostname and in part by a feature self-similarity matrix computed for each group. The parameters of the representation (histogram bins) are optimized and learned based on the training samples along with the classifiers. The proposed classification system was deployed on large corporate networks, where it detected 2,090 new and unseen variants of malware samples with 90% precision (9 of 10 alerts were malicious), which is a considerable improvement when compared to the current flow-based approaches or existing signaturebased web security devices.
