Decentralized/Federated Learning

We take an innovative approach to design and implement a purely decentralized machine learning system; the implementation includes algorithms, simulation systems, and middleware. In addition, the research work has been extended to designing a blockchain-based incentive system.

Yu, Haoxiang, Hsiao-Yuan Chen, Sangsu Lee, Xi Zheng, and Christine Julien. “Prototyping Opportunistic Learning in Resource-Constrained Mobile Devices” Proceedings of the 1st Workshop on Pervasive and Resource-Constrained Artificial Intelligence (at PerCom). 2022

Lee, Sangsu, Haoxiang Yu, Xi Zheng, and Christine Julien. “Swarm: Playground for Large-scale Decentralized Learning Simulations” Proceedings of the IEEE Pervasive Computing and Communication (Demo Paper). 2022