Amina Bassit - ESR9
Amina Bassit received her BSc degree in discrete mathematics from Hassan II University, Morocco, in 2013. Subsequently, she earned her MSc degree in cryptography at both Mohammed V University, Morocco, and Limoges University, France, in the fall of 2015. From 2016 to 2019, she was employed as a cryptographer at DGSSI, Morocco. Since April 2019, Amina has joined the Data Management and Biometrics (DMB) group and the Services and Cyber-Security (SCS) group of University of Twente, The Netherlands, to design a maliciously secure biometric verification protocol in the encrypted domain. Then from May 2020, she started working on her PhD as part of the PriMa (Privacy Matters) project where she is investigating the homomorphically encrypted likelihood-ratio-based (HELR) classifier in deep-learning-based biometric recognition; along with maliciously secure multi-party protocols.
Biometric recognition, deep learning, homomorphic encryption, secure protocol design, provable security, multi-part computation, malicious model.
ESR9: Integration of biometric recognition and homomorphic encryption
Recent research has shown that a novel approach integrating the optimal likelihood-ratio-based classifier in a homomorphic encryption scheme results in a very fast biometric recognition under encryption with near optimal recognition performance. The specific research topic of this position is to develop this approach further. The researcher will investigate the integration of this approach in deep-learning-based biometric recognition and the extension to other multi-party schemes and more malicious attack scenarios.
Bjarte M. Østvold