Seminar: Opportunities and Limits of High-Accuracy Differentially Private Deep Learning
Date and time: 21 November 2023, 15:15 – 16:15 CET
Speaker: Antti Honkela, University of Helsinki
Title: Opportunities and Limits of High-Accuracy Differentially Private Deep Learning
Where: Digital Futures hub, Osquars Backe 5, floor 2 at KTH main campus or Zoom
Directions: https://www.digitalfutures.kth.se/contact/how-to-get-here/
Zoom: https://kth-se.zoom.us/j/62461202124
This seminar is co-sponsored by Digital Futures.
Abstract: Differential privacy (DP) is widely regarded as the gold standard privacy definition for machine learning and data analysis. The strong privacy protection can severely limit the accuracy of models trained under DP. Recent work has shown that the degradation of accuracy can be avoided by using fine-tuning of large pre-trained models. We explore the phenomenon further to understand its limits regarding the amount of data needed and the similarity of pre-training and target data. I will also review the building blocks needed for high-accuracy DP deep learning.
Bio: Antti Honkela is a Professor of Data Science (Machine Learning and AI) at the Department of Computer Science, University of Helsinki. He is the coordinating professor of the Research Programme in Privacy-preserving and Secure AI at the Finnish Center for Artificial Intelligence (FCAI), a flagship of research excellence appointed by the Academy of Finland, and leader of the Privacy and infrastructures work package in European Lighthouse in Secure and Safe AI (ELSA), a European network of excellence in secure and safe AI. He serves in multiple advisory positions for the Finnish government in the privacy of health data.