Anto Joseph

Talk Title

Adversarial Machine Learning

Abstract

Machine Learning is not the future , systems we use today,use ML extensively , be it an e-com website or fraud detection in banking.It takes some skill and experience to apply machine learning and get useful results, especially with little research into the application of ML in Infosec. But it's a skill that anyone can learn! We start out with some basic boolean classification problem and introduce classifiers. Classifiers could be considered the brain of your ML system. We deal with some easy to learn classifiers in scikit learn , like a linear classifiers, Decision Trees etc and visualize how it works. We dive into training our classifiers with a labelled dataset.Trying different classifiers to approach the problem and verify the accuracy by cross verifying with the test data helps us choose an ideal algorithm for the problem in hand. This lab servers as a quick and practical introduction to the world of machine learning.

Speaker Bio

Anto Joseph is a Security Engineer for Intel. He has 4 years of corporate experience in developing and advocating security in Mobile and Web Platforms. Machine Learning systems is one of his key areas on Interest . He is very passionate about exploring new ideas in these areas and has been a presenter and trainer at various security conferences including BH USA 2016, Defcon 24, BruCon , HackInParis, HITB Amsterdam , NullCon , GroundZero , c0c0n , XorConf and more. He is an active contributor to many open-source projects and some of his work is available at https://github.com/antojoseph