M.I. Gregoriadis

M.I. Gregoriadis is a Researcher at the Faculty of Electrical Engineering, Mathematics and Computer Science at Delft University of Technology (TU Delft). His work focuses on Machine Learning, Decentralized Systems, and Data Engineering, with notable contributions to federated learning, adaptive ranking algorithms, and blockchain-based data analysis.

• Key Research Areas: Machine Learning, Ranking Algorithms, User Behavior, Data Deduplication
• Collaborations: ACM, 4TU.ResearchData

Recent work includes decentralized search engines (SwarmSearch), differentiable search indexes (De-DSI), and federated learning datasets. His publications address technical challenges in content-defined chunking, IPFS deduplication, and blockchain analytics.

Gregoriadis co-created the Tribler Learning-to-Rank Dataset and collaborates with researchers like Q. Stokkink and J. Pouwelse. His work emphasizes practical implementations in decentralized systems and open-access research outputs.