This paper investigates multiple parallel federated learning in cellular networks, where a base station schedules several FL tasks and each task has group of devices involved. To reduce the communication overhead, over-the-air computation is introduced by utilizing superposition property access channels (MAC) to accomplish aggregation step. Since all use same radio resource transfer their local...

A performance prediction method for massively parallel computation is proposed. The based on modeling and Bayesian inference to predict elapsed time T as a function of the number used nodes P (T=T(P)). focus extrapolation larger values from perspective application researchers. proposed has several improvements over developed in previous paper, real-symmetric generalized eigenvalue problem shows...