Mohamed Maouche

Mohamed Maouche is a Research Scientist (equivalent to tenured Associate Professor) at Inria's Privatics Team, specializing in privacy-preserving machine learning. His work focuses on developing anonymization techniques for diverse data types including speech, mobility, and health data, with emphasis on federated learning and differential privacy frameworks.

Education

• PhD in Computer Science, INSA Lyon, France (2016–2019)
• Computer Science Engineer, Ecole Nationale Superieure d’Informatique, Algiers (2011–2016)
• Baccalaureat in Mathematics, Bouamama High School, Algiers (2008–2011)

Research Focus

Maouche's research addresses privacy-utility tradeoffs in machine learning systems. Key areas include:

• Anonymization techniques for speech, mobility, and medical data
• Re-identification threat analysis across applications
• Privacy-preserving federated learning for distributed systems
• Differential privacy implementations

His publications demonstrate consistent focus on practical privacy solutions, with recent emphasis on synthetic data generation and speech anonymization. Earlier work established foundational methods for location privacy protection.

Teaching Experience

• Dimension Reduction (Master's level, Université de Lille)
• Computer Science fundamentals (INSA-Lyon, 2016–2019)
• Systems programming, Web Data, and HCI (INSA-Lyon)