Manan Mehta
Papers
1
Total Citations
56
H-Index
1
About
Manan Mehta is a rising researcher in distributed machine learning, with a primary focus on federated learning and its real-world deployment across heterogeneous systems. His most influential work, "A Greedy Agglomerative Framework for Clustered Federated Learning" (2023), has already garnered 56 citations, addressing a critical bottleneck in FL: the challenge of training models across non-IID, multi-source data typical of healthcare, smart manufacturing, autonomous driving, and robotics. Mehta’s key contribution lies in developing a greedy agglomerative clustering algorithm that dynamically groups clients with similar data distributions, significantly improving model accuracy and convergence speed without compromising privacy. This framework offers a practical, scalable solution for industrial big data environments where data silos and statistical heterogeneity are the norm. By bridging the gap between theoretical FL advances and applied systems, Mehta’s work is shaping how decentralized intelligence can be efficiently harnessed in privacy-sensitive sectors. His research continues to explore robust aggregation strategies and communication-efficient protocols, positioning him as a promising voice in the next wave of federated learning innovation.
Research Focus
Key Achievements
Top Papers
- 1A Greedy Agglomerative Framework for Clustered Federated Learning56 citations · 2023