Hi there! 👋 I'm Reda Bellafqira

👨‍🏫 Associate Professor | 🔬 AI Security Researcher | 🤖 AI/ML Enthusiast

Welcome to my GitHub profile! I'm an Associate Professor at IMT Atlantique, specializing in federated learning, deep neural network watermarking, and cryptography. My work focuses on developing secure and robust AI systems, particularly in sensitive domains like healthcare.

---

🚀 About Me

• 👨‍🏫 Position: Associate Professor at IMT Atlantique
• 🔬 Research Focus: Federated Learning, DNN Watermarking, AI Security, Healthcare Data Privacy
• 📚 Teaching: C++ Programming, Cryptography
• 🏛️ Institution: IMT Atlantique, France
• 📧 Contact: reda.bellafqira@imt-atlantique.fr
• 🌍 Based in: France
• 📊 Publications: 35+ research papers with 239+ citations

🔬 Research Interests

🔬 Research Areas:
├── AI Security
│   ├── Deep Neural Network Watermarking
│   └── Malicious Model Detection
├── Privacy Tech
│   ├── Federated Learning
│   └── Healthcare Data De-identification
├── Applications
│   └── Medical AI
└── Security
    ├── Cryptography
    ├── Watermarking
    └── Blockchain

📚 Teaching Areas:
├── Programming
│   ├── C++ Development
│   ├── Design Patterns
│   └── UML Diagrams
└── Security
    └── Applied Cryptography

🎓 Teaching & Education

As an Associate Professor at IMT Atlantique, I'm passionate about education and mentoring the next generation of cybersecurity and AI professionals:

📚 Courses I Teach:

• C++ Programming - Advanced programming concepts, data structures, and algorithms
• Cryptography - Classical and modern cryptographic techniques, security protocols

🎯 Research Supervision:

• PhD students and research engineers in AI Security, Federated Learning, Homomorphic Encryption, Secure Multiparty Computation, Image watermarking
• Master's thesis projects in privacy-preserving machine learning
• International collaborations and research exchanges

---

🔬 Featured Research Projects

🔐 DICTION - DNN Watermarking

DICTION: DynamIC robusT whIte bOx Watermarking Scheme for Deep Neural Networks

• 🎯 Protecting intellectual property of deep neural networks in white-box settings
• 🛡️ Robust against various attacks and model modifications
• 📄 Research Paper

🔍 FedCAM - Malicious Model Detection

FedCAM: Identifying Malicious Models in Federated Learning Environments Conditionally to Their Activation Maps

• 🕵️ Detects malicious participants in federated learning environments
• 🧠 Uses activation maps for conditional identification to detect Byzantine and backdooring attacks
• 🏥 Applications in healthcare and sensitive domains
• 📄 Research Paper

📊 Histogram Shifting Predictions

A Blockchain-Enhanced Reversible Watermarking Framework for End-to-End Data Traceability in Federated Learning Systems

• 📈 Statistical analysis and prediction methods
• 🔬 Research-grade implementation
• 📊 Data visualization and processing tools
• 📄 Research Paper

📚 Recent Publications

Selected Recent Work:

• "DICTION: DynamIC robusT whIte bOx Watermarking Scheme for Deep Neural Networks"
• "FedCAM: Identifying Malicious Models in Federated Learning Environments Conditionally to Their Activation Maps"
• "Automatic Deidentification of French Electronic Health Records: A Cost-Effective Approach Exploiting Distant Supervision and Deep Learning Models"
• "When Federated Learning Meets Watermarking: A Comprehensive Overview of Techniques for Intellectual Property Protection"
• "A Blockchain-Enhanced Reversible Watermarking Framework for End-to-End Data Traceability in Federated Learning Systems"
• "Secure Extraction of Personal Information from EHR by Federated Machine Learning"
• "FedCrypt: A Dynamic White-Box Watermarking Scheme for Homomorphic Federated Learning"

📖 View Full Publication List

🛠️ Technical Skills

Languages & Frameworks

Research & Security

📊 GitHub Statistics

🌟 Research Impact

📊 Research Metrics:
├── 📄 Publications: 35+
├── 📈 Citations: 239+
├── 🔗 H-index: Growing
└── 🌍 International Collaborations

🎯 Current Research Focus

graph TD
    A[AI Security Research] --> B[Federated Learning]
    A --> C[DNN/Image Watermarking]
    A --> D[Healthcare Privacy]
    B --> E[Malicious Detection]
    B --> F[Robust Aggregation against Byzantine and Backdooring Attacks]
    C --> G[IP Protection]
    C --> H[Model Authentication]
    D --> I[EHR De-identification]
    D --> J[Privacy-Preserving ML]

Loading

🤝 Collaboration & Contact

I'm always interested in collaborating on:

• 🔬 Federated Learning research and applications
• 🛡️ DNN watermarking and robustness studies
• 🏥 Healthcare AI and privacy-preserving technologies
• 📚 Open Source projects in ML security

Let's Connect: 🤝

• 📧 Email: reda.bellafqira@imt-atlantique.fr
• 🔗 Google Scholar: My Publications
• 🌐 ResearchGate: My Profile

---

"Advancing AI security and privacy through innovative research and open collaboration"

⭐ If you find my work interesting, feel free to star my repositories! ⭐