Machine Learning for Cyber Security

Increasingly, cyber-attacks are sophisticated and occur rapidly, necessitating the use of machine learning techniques for detection at machine speed. However, the use of traditional machine learning techniques in cyber security requires subject matter expertise (i.e., network analysts) to extract relevant and distinctive features from the raw network traffic. Thus, we propose a novel machine learning algorithm for malicious network traffic detection using only the bytes of the raw network traffic. We also propose a transfer learning architecture to enable training and inference, respectively in a source and target network environment.