Workshop
Introduction to Machine Learning
Abstract:
Covers machine learning functionality, attacks and defenses. We'll
attack public Large Learning Models with prompt injection, and make
custom machine learning models with Python. We'll create various
models including linear regression, polynomial regression, and Support
Vector Machines, train them, and evaluate their performance. Projects
include computer vision, breaking a CAPTCHA, deblurring images,
regression, and classification tasks. We will perform poisoning and
evasion attacks on machine learning systems, and implement deep neural
rejection to block such attacks.
No experience with programming or machine learning is required, and
the only software required is a Web browser. We will use TensorFlow
and SecML on free Google Colab cloud systems.
Bio:
Sam Bowne has been teaching computer networking and security classes at City College San Francisco since 2000. He has given talks and hands-on trainings at DEF CON, DEF CON China, Black Hat USA, HOPE, BSidesSF, BSidesLV, RSA, and many other conferences and colleges. He founded Infosec Decoded, Inc., and does corporate training and consulting for several Fortune 100 companies, on topics including Incident Response and Secure Coding.
Panel:
Diane Lin -
Joe Hall -
Malcom Harkins -