Dominik Fay

PhD student at KTH, Research scientist at Elekta. 📍 Stockholm, Sweden.

prof_pic2.jpg

⚠️🚧 This website is under construction 🚧⚠️

Hi! I’m an industrial PhD student at KTH and Elekta advised by Mikael Johansson. I work on problems in machine learning and data privacy and applications in radiation therapy.

Previously, I did my MSc in Computer Science at KTH and my BSc in Computer Science at Heidelberg University. I also interned at Chalmers and EMBL.

news

Dec 14, 2023 Our paper on dynamic privacy allocation for federated learning has been accepted to ICASSP 2024! :tada: You can find the preprint here. Congrats to the lead author Jiaojiao and to Mikael! :sparkles:
Nov 21, 2023 I have successfully completed my licentiate degree! :tada: You can find my thesis here.
Feb 17, 2023 My first patent has been granted :tada: It was under review for more than 3 years.

selected publications

  1. Privacy Amplification via Importance Sampling
    Dominik Fay, Sebastian Mair, and Jens Sjölund
    CoRR, 2023
    PDF
  2. Adaptive Hyperparameter Selection for Differentially Private Gradient Descent
    Dominik Fay, Sindri Magnússon, Jens Sjölund, and 1 more author
    Transactions on Machine Learning Research, 2023
    PDF
  3. FDR-controlled metabolite annotation for high-resolution imaging mass spectrometry
    Andrew Palmer, Prasad Phapale, Ilya Chernyavsky, and 8 more authors
    Nature methods, 2017