Title: 

Machine Learning and Privacy: Challenges and Opportunities

Abstract:

The emergence of powerful machine learning methods presents both challenges and opportunities for data privacy research.  On the one hand, machine learning models trained on sensitive data present new privacy risks and open the door to new types of inference attacks. On the other hand, many objectives of modern machine learning - in particular, constructing generalizable models that are not overfitted to the training data - are compatible with privacy and benefit from the same set of techniques.  In this talk, I will discuss several open research questions at the junction of machine learning and privacy.