Current status 🏫
In October 2021, I started a PhD in Federated Learning between Inria Lille and the University of Twente. I am supervised by Jan Ramon (Inria Lille), Florian Hahn (UTwente) and Andreas Peter (UTwente/University of Oldenburg).
The goal of my thesis is to provide highly scalable and secure solutions for Federated Learning with a particular focus on resource-constrained devices (e.g. smartphones).
Research Interests 🔬
I have two main research interests are:
- Federated Learning, especially its security aspect. Hence, I read papers about SMPC, Homomorphic Encryption, Function Secret Sharing, etc. My intent isn’t to build highly complex ML systems but simply to propose generic secure building blocks for them.
- Searchable Encryption. This is a subject on which I worked during a previous research internship with Florian Hahn and Andreas Peter. I still have some results and ideas that will result in occasional publications on this subject.
More generally, my overall interest is in the privacy-preserving use of data.
Besides these technical interests, I am open to many other topics. Especially, I am looking forward to any insights on ethics or environmental evaluation of computer systems. These elements essentially contribute to my personal reflections about my long-term role as a computer scientist in world facing multiple crisis.
- M. Dijsklag, M. Damie, F. Hahn and A. Peter, ‘Passive query-recovery attack against secure conjunctive keyword search schemes’, ACNS 2022 [to be published] (B)
- M. Damie, F. Hahn, and A. Peter, ‘A Highly Accurate Query-Recovery Attack against Searchable Encryption using Non-Indexed Documents’, USENIX Security 21 (A*)
- Sub-reviews: PETS 2023, CODASPY 2022
- Program committee: NDSS Artifact Evaluation 2024, PETS Artifact Evaluation 2024
- Nomination for the Dutch CyberSecurity Research Paper (i.e. top 3) with the paper ‘A Highly Accurate Query-Recovery Attack against Searchable Encryption using Non-Indexed Documents’