The deluge of information security data invites data-driven strategies for situational awareness, threat detection and remediation.  Statistics and machine learning approaches to assess and automate elements of information security have become increasingly popular. However, there exist few venues for collegial information exchange in the level of technical detail appropriate for data science practitioners. The Conference on Applied Machine Learning for Information Security (CAMLIS) provides a venue for discussing applied work from researchers in academia, government research labs, national laboratories and FFRDCs, and information security data scientists in the industry.

 
 

Topics may include but are not limited to:
    •    Insider threat detection
    •    Network and endpoint forensics
    •    Governance, compliance and exfiltration detection
    •    Detection of script-based and malware-less attacks
    •    Automated malware detection and classification
    •    Vulnerability assessment

ML techniques and analytic or predictive themes might include:
    •    Statistical analysis on large and small datasets
    •    Unique considerations of base-rate fallacy for data science in information security
    •    Data sources and data exploration and subsequent findings
    •    Unique approaches to data visualization
    •    Unsupervised methods and anomaly detection
    •    Adversarial machine learning
    •    Original or cross-domain deep learning architectures applied to information security data
    •    Natural language processing
    •    Reinforcement learning for automating security tasks