Antti Honkela
Professor of Data Science, University of Helsinki
Workshop 1
Differential Privacy and Unlearning in Machine Learning
Overview
Workshop 1 is part of the Data Privacy in Machine Learning P1 program. The workshop will bring together tutorials, invited talks, and contributed talks on differential privacy, machine unlearning, and privacy-preserving machine learning.
The program includes tutorials on differential privacy and machine unlearning, invited talks by Antti Honkela, Graham Cormode, and Anastasiia Koloskova, and contributed talks selected through the call for contributions.
Registration
Registration deadline: Sunday, June 7, 2026
Speakers
Graham Cormode
Professor of Computer Science, University of Oxford
Anastasiia Koloskova
Assistant Professor of AI and Optimization, UZH
Call for Contributions
We are seeking submissions for 15-20 minute talks on topics related to privacy-preserving machine learning and unlearning. The workshop will not have a poster session.
Topics of interest include:
- Theory and applications of differential privacy
- Theory and applications of machine unlearning
- Relaxations of differential privacy, and relation to other privacy notions and methods
- Unlearning in large language models
- Attacks and auditing of privacy and unlearning
- Any other topic related to the core theme of the workshop
Registration formRegister for the workshop by Sunday, June 7, 2026
Preliminary Agenda
Tutorial: Introduction to Differential PrivacyRasmus Pagh
Lunch break
Invited TalkOn privacy promises - when is my algorithm (differentially) private enough?Antti Honkela
Near-Optimal Generalized Private TestingAnamay Chaturvedi
Modular Verification of Differential Privacy in Probabilistic Higher-Order Separation LogicAlejandro Aguirre
Coffee break
Invited TalkDifferentially Private Analytics for Federated ApplicationsGraham Cormode
Local Node Differential PrivacyAnatoly Zavyalov
Privacy–Robustness Trade-offs in Federated Learning: From Optimization to GeneralizationNirupam Gupta
Coffee and pastries
Differentially Private Modeling of Disease Transmission within Human Contact NetworksDebanuj Nayak
Tutorial: Introduction to Machine UnlearningAmartya Sanyal
Invited TalkCertifiable Algorithms and Auditing Methods for Machine UnlearningAnastasiia Koloskova
Lunch break
Why Forget only Unlearning needs MemorizationVikrant Singhal
Less Noise, Same Certificate: Retain Sensitivity for UnlearningCarolin Heinzler
End of workshop
Venue
Lokale: Aud 02 AKB, Universitetsparken 13
Open in OpenStreetMap · Map data © OpenStreetMap contributors
Related Events in Copenhagen
- CPM 2026 Summer School on Strings & Privacy, June 13-14, 2026
- Learning Theory Summer School and Workshop Copenhagen, June 22-27, 2026