Workshop on Trustworthy Learning
Date and time: 16 October 2023, 08:30-17:30 CEST (UTC +2)
Title: Workshop on Trustworthy Learning
Where: Digital Futures hub, Osquars Backe 5, floor 2 at KTH main campus OR Zoom
A maximum of 50 participants are onsite at Digital Futures cafeteria. First come, first serve.
For participation, please register here: https://www.kth.se/form/64e8b6f3917ea4c31c2d46a7
If you are unable to participate ONSITE, you are welcome to join us ONLINE. A Zoom link will be sent a few days before the event to those who have signed up for online participation.
Welcome to this full-day workshop on trustworthy learning, emphasising data privacy and integrity, explainable machine learning, and the cross-disciplinary challenges between law and technology. Stimulating discussions and educational presentations should help to bring professionals from both disciplines closer.
The workshop is organised by the two collaborative projects DataLEASH and EXTREMUM funded by Digital Futures.
09:10 Plenary talk: An approach to reconciling computer science and legal approaches to privacy – by Kobbi Nissim, Professor at the Department of Computer Science, Georgetown University and an Affiliate Professor at Georgetown Law – including Q&A
- Sara Saeidian and Tobias Oechtering, Pointwise Maximal Leakage – A novel privacy measure for qualitative privacy risk assessment
- Hercules Dalianis, Protecting Privacy in the Age of Large Language Models
10:40 COFFEE BREAK
- Stanley Greenstein and Rami Mochaourab, Embedding Legal Values into AI
- Sonja Buchegger, Synthetic data for privacy-preserving ML
- Per Nyden, Legal Specialist, Swedish Authority for Privacy Protection (IMY)
- Annika Palm, Expert, Government Offices of Sweden
- Helena Haapio (remote), Associate Professor of Business Law, University of Vaasa
- Helene Spjuth, Head of Centre for Health data, Region Stockholm
12:30 LUNCH BREAK & POSTER session with more technical details
13:30 Plenary talk Nicklas Berild Lundblad, Head of Public Policy and Public Affairs DeepMind – remote including Q&A
- Panagiotis Papapetrou, Explainable Machine Learning for Healthcare: Challenges and future Outlook
- Ioanna Miliou, Explaining temporal sequences
15:00 COFFEE BREAK
- Sepideh Pashami, Adversarial Robust Machine Learning
- Cristian Rojas, The Two-Stage Method: A computationally efficient and private approach to statistical learning
15:50 Group discussion on future challenges and development
16:30 POSTER session & mingle
17:30 End of Day