Cyber-Informed Machine Learning: End-User Value through Explainability


Posted on in Presentations

Cybersecurity is a data-rich and natural setting for machine learning (ML). However, lack of explainability is a major challenge. This session will propose cyber-informed ML, a paradigm emphasizing two directions of explainability, human-to-model and model-to-human. Will share practical examples of overcoming this challenge and discuss research needed for ML at the cybersecurity operations level.

Participants
Jeffrey Mellon

Speaker

Machine Learning Research Scientist, Carnegie Mellon University Software Engineering Institute

Clarence Worrell

Speaker

Senior Data Scientist, Carnegie Mellon University Software Engineering Institute


Share With Your Community