Federated Learning Survey: A Multi-Level Taxonomy of Aggregation Techniques, Experimental Insights, and Future Frontiers DOI
Meriem Arbaoui, Mohamed-El-Amine Brahmia, Abdellatif Rahmoun

et al.

ACM Transactions on Intelligent Systems and Technology, Journal Year: 2024, Volume and Issue: 15(6), P. 1 - 69

Published: July 17, 2024

The emerging integration of Internet Things (IoT) and AI has unlocked numerous opportunities for innovation across diverse industries. However, growing privacy concerns data isolation issues have inhibited this promising advancement. Unfortunately, traditional centralized Machine Learning (ML) methods demonstrated their limitations in addressing these hurdles. In response to ever-evolving landscape, Federated (FL) surfaced as a cutting-edge ML paradigm, enabling collaborative training decentralized devices. FL allows users jointly construct models without sharing local raw data, ensuring privacy, network scalability, minimal transfer. One essential aspect revolves around proficient knowledge aggregation within heterogeneous environment. Yet, the inherent characteristics amplified complexity its practical implementation compared ML. This survey delves into three prominent clusters research contributions: personalization, optimization, robustness. objective is provide well-structured fine-grained classification scheme related areas through unique methodology selecting work. Unlike other papers, we employed hybrid approach that amalgamates bibliometric analysis systematic scrutinizing find most influential work literature. Therefore, examine challenges contemporary techniques heterogeneity, efficiency, security, privacy. Another valuable asset study comprehensive coverage strategies, encompassing architectural features, synchronization methods, several federation motivations. To further enrich our investigation, insights evaluating novel proposals conduct experiments assess compare under IID non-IID distributions. Finally, present compelling set avenues call exploration open up treasure

Language: Английский

Federated Learning Survey: A Multi-Level Taxonomy of Aggregation Techniques, Experimental Insights, and Future Frontiers DOI
Meriem Arbaoui, Mohamed-El-Amine Brahmia, Abdellatif Rahmoun

et al.

ACM Transactions on Intelligent Systems and Technology, Journal Year: 2024, Volume and Issue: 15(6), P. 1 - 69

Published: July 17, 2024

The emerging integration of Internet Things (IoT) and AI has unlocked numerous opportunities for innovation across diverse industries. However, growing privacy concerns data isolation issues have inhibited this promising advancement. Unfortunately, traditional centralized Machine Learning (ML) methods demonstrated their limitations in addressing these hurdles. In response to ever-evolving landscape, Federated (FL) surfaced as a cutting-edge ML paradigm, enabling collaborative training decentralized devices. FL allows users jointly construct models without sharing local raw data, ensuring privacy, network scalability, minimal transfer. One essential aspect revolves around proficient knowledge aggregation within heterogeneous environment. Yet, the inherent characteristics amplified complexity its practical implementation compared ML. This survey delves into three prominent clusters research contributions: personalization, optimization, robustness. objective is provide well-structured fine-grained classification scheme related areas through unique methodology selecting work. Unlike other papers, we employed hybrid approach that amalgamates bibliometric analysis systematic scrutinizing find most influential work literature. Therefore, examine challenges contemporary techniques heterogeneity, efficiency, security, privacy. Another valuable asset study comprehensive coverage strategies, encompassing architectural features, synchronization methods, several federation motivations. To further enrich our investigation, insights evaluating novel proposals conduct experiments assess compare under IID non-IID distributions. Finally, present compelling set avenues call exploration open up treasure

Language: Английский

Citations

0