Published in IMWUT (UbiComp), 2022
Introduces PEPPER, a decentralized gossip-based framework for recommender systems.
Recommended citation: Yacine Belal, Aurélien Bellet, Sonia Ben Mokhtar, Vlad Nitu. (2022). "PEPPER." Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., 6(3), 101:1–101:27.
Download Paper
Published in ACM Transactions on Spatial Algorithms and Systems, 2024
A comprehensive survey of federated learning models applied to spatial-temporal mobility contexts.
Recommended citation: Yacine Belal, Sonia Ben Mokhtar, Hamed Haddadi, Jaron Wang, Afra Mashhadi. (2024). "Survey of Federated Learning Models for Spatial-Temporal Mobility Applications." ACM Transactions on Spatial Algorithms and Systems, 10(3), 1–39.
Download Paper
Published in arXiv, 2024
Analyzes privacy risks in decentralized learning from a membership inference standpoint.
Recommended citation: Ousmane Touat, Jezekael Brunon, Yacine Belal, Julien Nicolas, Mohamed Maouche, César Sabater, Sonia Ben Mokhtar. (2024). "Scrutinizing the Vulnerability of Decentralized Learning to Membership Inference Attacks." arXiv preprint arXiv:2412.12837.
Download Paper
Published in arXiv, 2025
Introduces GRANITE, a framework that encompasses a history-based Random Peer Sampling protocol and a higher bound on the byzantine presence which is leveraged by underying robust aggregators.
Recommended citation: Yacine Belal, Mohamed Maouche, Sonia Ben Mokhtar, Anthony Simonet-Boulogne. (2025). "GRANITE: a Byzantine-Resilient Dynamic Gossip Learning Framework." arXiv preprint arXiv:2504.17471.
Download Paper
Published in IEEE DCOSS-IoT (To Appear), 2025
Benchmarks clustered FL algorithms for next-location prediction in smart systems.
Recommended citation: Sonal Yadav, Yacine Belal, Brent Lagesse, Afra Mashhadi. (2025). "Benchmarking Clustered Federated Learning Algorithms for Next-Point Prediction." IEEE DCOSS-IoT. To appear.
Download Paper
Published in IEEE ICDCS (To Appear), 2025
A privacy attack that aims at identifying communities of interest in collaborative recommender systems.
Recommended citation: Yacine Belal, Mohamed Maouche, Sonia Ben Mokhtar, Anthony Simonet-Boulogne. (2025). "Inferring Communities of Interest in Collaborative Learning-based Recommender Systems." IEEE International Conference on Distributed Computing Systems (ICDCS). To appear.
Download Paper
Published:
Presented PEPPER at UbiComp 2022, introducing a decentralized framework for recommender systems.
Published:
Presented recent research on privacy-preserving recommender systems and federated learning.
Published:
Invited talk on Byzantine resilience techniques in decentralized gossip learning systems.
Published:
Presented work on Byzantine-resilient gossip learning in dynamic and sparse settings.
Course, INSA Lyon, 2021
Delivered 24 hours of lab sessions to fifth-year engineering students. Covered: distributed architectures, federated learning, differential privacy, and data heterogeneity.
Course, INSA Lyon, 2022
Delivered 40 hours of lab sessions to fifth-year engineering students. Covered: web scraping, data storage and processing in AWS, and data analysis/visualization with pandas.
Course, INSA Lyon, 2023
Delivered 20 hours of lab sessions to 5th-year engineering students in 2024. Covered concepts and tools for real-time data processing, including stream models, windowing, and frameworks like Apache Kafka and Flink.
Workshop, INSA Lyon, 2023
Delivered 48 hours of foundational mathematics lectures across the 2023 and 2024 summer sessions for incoming first-year engineering students. Covered core topics in Linear Algebra, Calculus, and Probability.
Course, INSA Lyon, 2024
Delivered 48 hours of lab sessions to 4th-year engineering students starting from October 2024. Topics included socket programming, client-server architectures, and practical network development in C.