TY  - RPRT
TI  - Differentially private training of neural networks with Langevin dynamics for calibrated predictive uncertainty
AU  - Moritz Knolle
AU  - Alexander Ziller
AU  - Dmitrii Usynin
AU  - Rickmer Braren
AU  - Marcus R. Makowski
AU  - Daniel Rueckert
AU  - Georgios Kaissis
PY  - 2021
UR  - https://arxiv.org/abs/2107.04296
ID  - 2107.04296
ER  -