Adaptively Fingerprinting Users/Applications at Scale with GPU Acceleration


Posted on in Presentations

The massive scale, dimensionality, and heterogeneity of user application data requires an adaptive approach to detect malicious actions. This session will demo Morpheus, an open source framework, that supports assembling of multiple machine learning models to dynamically fingerprint users based on time series, categorical, and quantitative data and extract deviations from typical activity.


Participants
Rachel Allen

Speaker

Senior Data Scientist, Cybersecurity, Nvidia

Gorkem Batmaz

Speaker

Senior Data Scientist, Cybersecurity, Nvidia

Arjun Chakraborty

Speaker

Senior Software Engineer, Security Analytics, Nvidia

DevSecOps & Application Security Machine Learning & Artificial Intelligence Open Source Tools

application security cloud security insider threats intrusion prevention/detection


Topic

Subtopic


Share With Your Community