Aarhus University Seal / Aarhus Universitets segl

Special talk by Florian Tramer on Measuring and Enhancing the Security of Machine Learning

2021.02.15 | Søs Küster Markussen

Date Thu 11 Mar
Time 17:00 18:00
Location Online - Zoom meeting

Special talk by Florian Tramer on Measuring and Enhancing the Security of Machine Learning

Abstract:

Failures of machine learning systems can threaten both the security and privacy of their users. My research studies these failures from an adversarial perspective, by building new attacks that highlight critical vulnerabilities in the machine learning pipeline, and designing new defenses that protect users against identified threats. In the first part of this talk, I'll explain why machine learning models are so vulnerable to adversarially chosen inputs. I'll show that many proposed defenses are ineffective and cannot protect models deployed in overtly adversarial settings, such as for content moderation on the Web. In the second part of the talk, I'll focus on the issue of data privacy in machine learning systems, and I'll demonstrate how to enhance privacy by combining techniques from cryptography, statistics, and computer security.

CS frontpage, Featured, Public/media