University of Cambridge > Talks.cam > Computer Laboratory Security Seminar > Am I a Member? Auditing Private Machine Learning

  Browse A-Z

Log in

University Account External (via Google)

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹī¸ Information

Am I a Member? Auditing Private Machine Learning

Download to your calendar using vCal

If you have a question about this talk, please contact Alexandre Pauwels.

Current privacy evaluations in machine learning (ML) rely predominantly on membership inference attacks to validate claims of differential privacy and machine unlearning. By framing ML regulation as a Principal-Agent problem, we demonstrate that regulators cannot rely on such attacks alone due to information asymmetry. This can lead to a false sense of privacy for individuals whose data is being analyzed. Consequently, we advocate for a paradigm shift from statistical auditing to algorithmic guarantees. We conclude on the role that cryptography will play for these algorithmic guarantees to be verifiable by third parties.

This talk is part of the Computer Laboratory Security Seminar series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Computer Lab, FW11 and Online (zoom link on this page) Webinar & LT2, Computer Laboratory, William Gates Building. in person & via zoom

Other talks

Title to be confirmed LCLU Annual Talk: The Early Aqueous Environment of Mars Inferred From Mission Lifetime Results by the Curiosity Rover at Gale Crater Climate, uncertainty, and the slippery measurement of Antarctic ice Coming soon Good and Evil Plasma Cells: Guardians & Saboteurs of Immunity Machine Learning for Spectral Geometry and Vice Versa
 

Š 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity