What it does This repository is a curated list of resources for applying machine learning to cyber security. It collects datasets, academic papers, books, conference talks, tutorials, and courses into a single index. Think of it as a bibliography with working hyperlinks.

The interesting bit The list doesn’t just cheerlead for ML; it includes talks like “Defeating Machine Learning” and papers on adversarial SVMs, acknowledging that attackers weaponize the same techniques. That self-awareness is rarer than it should be in security resource lists.

Key highlights

• Datasets from DARPA, LANL, NSA, and UNSW ADFA for intrusion detection research
• Papers covering DNS reputation systems (Notos, Kopis, EXPOSURE), malware detection (Polonium, Nazca), and password guessability via neural networks
• Talks range from Stanford’s “Data Mining for Cyber Security” course to a 5-minute “Build an Antivirus” video
• Includes adversarial perspectives: talks on pwning deep learning systems and defeating ML-based malware detection
• Single Stanford course (CS259d) listed under Courses; miscellaneous section points to MIT research on human-in-the-loop attack prediction

Caveats

• Several links point to blogspot/amzn.to shorteners and may decay; no archival copies provided
• “Awesom” badge in the header appears to be a typo that has persisted
• No code, no tools, no implementation guidance — purely a reading list

Verdict Worth bookmarking if you’re a researcher or practitioner building literacy at the ML-security boundary. Skip it if you need runnable code or structured learning paths; this is a starting point, not a syllabus.