Daniel Bernau portrait

Contact

E:
E:

| |

About Me

I'm a security and privacy architect at SAP in Walldorf, Germany. I hold a PhD in information security from the University of Stuttgart and a pan-european EIT master's degree from Technical University of Berlin and Université de Rennes 1. My current research interests are privacy enhancing technologies for machine learning and collaborative analytics, particularly anonymization with Differential Privacy.

Publications

D. Bernau, J. Robl, F. Kerschbaum
Assessing Differentially Private Variational Autoencoders under Membership Inference
36th IFIP WG 11.3 Conference on Data and Applications Security and Privacy (DBSEC), 2022.
D. Bernau, G. Eibl, P. W. Grassal, H. Keller, F. Kerschbaum
Quantifying identifiability to choose and audit epsilon in differentially private deep learning
48th International Conference on Very Large Data Bases (VLDB), 2022.
D. Bernau, J. Robl, P. W. Grassal, S. Schneider, F. Kerschbaum
Comparing local and central differential privacy using membership inference attacks
35th IFIP WG 11.3 Conference on Data and Applications Security and Privacy (DBSEC), 2021.
B. Hilprecht, M. Haerterich, D. Bernau
Monte Carlo and Reconstruction Attacks against Generative Models
Privacy Enhancing Technologies Symposium (PETS), 2019.
G. Eibl, K. Bao, P. Grassal, D.Bernau, H. Schmeck
The influence of Differential Privacy on short term electric load forecasting
7th DACH+ Conference on Energy Informatics, 2018.
J. Boehler, D. Bernau, F. Kerschbaum
Privacy-Preserving Outlier Detection for Data Streams
31st IFIP WG 11.3 Conference on Data and Applications Security and Privacy (DBSEC), 2017.

Invited Talks

CAST-Workshop hot topic: Kuenstliche Intelligenz und IT-Sicherheit, 2019
Talk: Quantifying and mitigating model leakage in deep learning with membership inference and differential privacy

SAP TechEd, 2019.
Talk: Enabling Privacy-Preserving Enterprise Applications

SAP Security Expert Summit, 2019.
Talk: Quantifying model leakage in deep learning with membership inference attacks

University of Applied Sciences Ludwigshafen, 2019.
Lecturer for Security and Privacy

SAP TechEd, 2018.
Talk: Data anonymization: unlocking the value in personal data

Funding