How can federated learning be extended and optimized for scenarios with limited communication bandwidth or high-latency networks, such as IoT devices or edge computing environments, while still preserving privacy and security?
Dear Sikander Hans,
I would suggest using the ideas of Digital Twins (DTs) for this. See the added Internet sources.
For general idea of DTs see my presentation at the address:
https://www.researchgate.net/publication/366167747_Digital_Twins_in_IoT
Best regards
Anatol Badach
The part „Use cases of a digital twin network“ in „Y.3090 : Digital twin network - Requirements and architecture“ at the address: https://www.itu.int/rec/T-REC-Y.3090-202202-I
Wei Yang; Wei Xiang; Yuan Yang; Peng Cheng: „Optimizing Federated Learning With Deep Reinforcement Learning for Digital Twin Empowered Industrial IoT“ https://ieeexplore.ieee.org/document/9815106
Kangde Liu, Zheng Yan, Xueqin Liang, Raimo Kantola, Chuangyue Hu: A survey on blockchain-enabled federated learning and its prospects with digital twin https://www.sciencedirect.com/science/article/pii/S2352864822001626
Selvarajan Shitharth, Hariprasath Manoharan , Achyut Shankar, et al.: Federated learning optimization: A computational blockchain process with offloading analysis to enhance security https://www.sciencedirect.com/science/article/pii/S1110866523000622
To extend and optimize federated learning for scenarios with limited communication bandwidth or high-latency networks, such as IoT devices, the following strategies can be considered:
By implementing these strategies, federated learning can be extended and optimized for scenarios with limited communication bandwidth or high-latency networks, such as IoT devices. These approaches aim to improve the efficiency, scalability, and performance of federated learning in constrained environments while ensuring data privacy and security.
- Badges
- Science topic
- Similar topics
- Filing