Sunday 3:00 PM–5:10 PM in C02

Machine Learning Security - The Data Scientist's Guide to Hardening ML Models

Arjun Bahuguna

Audience level:
Intermediate

Description

Your ML model is insecure. With increased attack incidents on machine learning models (adversarial images, membership inference, model inversion, information reconstruction, data poisoning, etc) it is essential for developers to understand the attack surface of their ML models. We will show how companies like Google & Microsoft are coping with these new threats, and how you can too.

Abstract

Outline:

Subscribe to Receive PyData Updates