Skip to yearly menu bar Skip to main content


Invited Talk

Robustness, Verification, Privacy: Addressing Machine Learning Adversaries

<div class="supplemental-html"> <ul style="list-style-type: none; line-height:1em; font-size:.9em; color:#666;padding: 5px !important;"> <li>Moderator: Avrim Blum </li> <li>On-demand video (45 minutes)</li> <li>Live Q&A (10 min)</li> <li>Break (5 min)</li> <li>Ask Me Anything Chat (up to an hour)</li> </ul> </div>

Shafi Goldwasser

Moderator : Avrim Blum


Abstract:

We will present cryptography inspired models and results to address three challenges that emerge when worst-case adversaries enter the machine learning landscape. These challenges include verification of machine learning models given limited access to good data, training at scale on private training data, and robustness against adversarial examples controlled by worst case adversaries.

Chat is not available.