Bachelor Thesis – Multi-Model Federated Learning
I worked with Prof. Sharayu Moharir at IIT Bombay to develop concept, framework and theoretical guarantees for simultaneous training of non-overlapping task in a Federated Learning set-up. We presented the idea, heuristic algorithms and the first proof-of-concept in the paper Multi-Model Federated Learning. Our subsequent paper Multi-Model Federated Learning with Provable Guarantees provides convergence guarantees for algorithms in our initial work.
- Bhuyan N. and Moharir S., “Multi-Model Federated Learning”, 2022 14th International Conference on
COMmunication Systems & NETworkS (COMSNETS), 2022, doi: 10.1109/COMSNETS53615.2022.9668435 - Bhuyan, N., Moharir, S., Joshi, G., “Multi-Model Federated Learning with Provable Guarantees”, VALUETOOLS
2022., Springer LNICST, volume 482, doi: 10.1007/978-3-031-31234-2_13
Research Internships
Synthetic Video Traffic Generation in DASH environment under Prof. Jimmy Nielsen at Aalborg University
Perception methods for Autonomous Vehicles in urban environments under Prof. Hayder Radha at Michigan State University