Associate Professor, IMT Atlantique — AI Security · Federated Learning · Applied Cryptography

I am an Associate Professor at IMT Atlantique (France), where I conduct research on the security and privacy of machine learning systems. My work focuses on deep neural network watermarking, robust federated learning, and privacy-preserving technologies for sensitive domains such as healthcare and genomics.

📧 reda.bellafqira@imt-atlantique.fr · Google Scholar · ORCID · DBLP · ResearchGate

• AI Security — Deep neural network watermarking (white-box and black-box), model authentication, detection of malicious models (Byzantine and backdoor attacks)
• Federated Learning — Robust aggregation, defense in non-IID environments, intellectual property protection of collaboratively trained models
• Privacy-Preserving Technologies — Homomorphic encryption, secure multiparty computation, de-identification of electronic health records
• Healthcare & Genomic Data Security — Privacy-preserving GWAS, secure DNA data storage, watermarking for data traceability

2025

• DICTION: DynamIC robusT whIte bOx Watermarking Scheme for Deep Neural Networks — Applied Sciences, 15(13)
• RoSe-Mix: Robust and Secure Deep Neural Network Watermarking in Black-Box Settings via Image Mixup — Machine Learning and Knowledge Extraction, 7(2)
• A Blockchain-Enhanced Reversible Watermarking Framework for End-to-End Data Traceability in Federated Learning Systems — CSP 2025
• A Dynamic Sliding Window Encoding for Secured DNA Data Storage Compliant With Biological and Indexing Constraints — IEEE Access, 13
• FedCLEAN: Byzantine Defense by Clustering Errors of Activation Maps in Non-IID Federated Learning Environments — arXiv preprint

2024

• FedCrypt: A Dynamic White-Box Watermarking Scheme for Homomorphic Federated Learning
• A White-Box Watermarking Modulation for Encrypted DNN in Homomorphic Federated Learning — SECRYPT 2024
• FedCAM: Identifying Malicious Models in Federated Learning Environments Conditionally to Their Activation Maps — WONS 2024
• Automatic De-identification of French Electronic Health Records: A Cost-Effective Approach Exploiting Distant Supervision and Deep Learning Models — BMC Medical Informatics and Decision Making, 24(1)
• Secure Extraction of Personal Information from EHR by Federated Machine Learning — MIE 2024

2023 and earlier (selection)

• When Federated Learning Meets Watermarking: A Comprehensive Overview of Techniques for Intellectual Property Protection — Machine Learning and Knowledge Extraction, 5(4), 2023
• Robust and Imperceptible Watermarking Scheme for GWAS Data Traceability — IWDW 2022
• Cryptosystem Conversion, Packing and Matrix Processing of Homomorphically Encrypted Data: Application to IoT Devices — IEEE Access, 9, 2021
• Privacy-Preserving Genome-Wide Association Study for Rare Mutations — IEEE Access, 8, 2020

35+ peer-reviewed publications — Full list on Google Scholar

• C++ Programming — Advanced programming, data structures, design patterns, UML
• Cryptography — Classical and modern cryptographic techniques, security protocols

I supervise PhD students, research engineers, and Master's theses on topics including AI security, federated learning, homomorphic encryption, secure multiparty computation, and image watermarking. I welcome inquiries from motivated students and potential collaborators.

I am open to academic and industrial collaborations on:

• Federated learning security and robustness
• Watermarking of machine learning models and multimedia content
• Privacy-preserving machine learning for healthcare and genomics

_{Python · C++ · PyTorch · TensorFlow · scikit-learn}