Decentralized descent optimization with stochastic gradient signs for device-to-device networks
TT Phuong - IEEE Wireless Communications Letters, 2021 - ieeexplore.ieee.org
IEEE Wireless Communications Letters, 2021•ieeexplore.ieee.org
We propose an algorithm for decentralized optimization in wireless device-to-device (D2D)
networks of pervasive devices such as sensors or 5G handsets, in which the signs of
stochastic gradient are used for descent steps. Our algorithm has the convergence rate of
(1/(nT)) in which is the number of devices and is the number of learning iterations, saving the
communication efficiency by at least 64 times when compared with previous results, and
being relatively robust to unexpected errors of adversarial scaling in communication …
networks of pervasive devices such as sensors or 5G handsets, in which the signs of
stochastic gradient are used for descent steps. Our algorithm has the convergence rate of
(1/(nT)) in which is the number of devices and is the number of learning iterations, saving the
communication efficiency by at least 64 times when compared with previous results, and
being relatively robust to unexpected errors of adversarial scaling in communication …
We propose an algorithm for decentralized optimization in wireless device-to-device (D2D) networks of pervasive devices such as sensors or 5G handsets, in which the signs of stochastic gradient are used for descent steps. Our algorithm has the convergence rate of (1/( nT )) in which is the number of devices and is the number of learning iterations, saving the communication efficiency by at least 64 times when compared with previous results, and being relatively robust to unexpected errors of adversarial scaling in communication. Theoretical claims are verified by numerical results on a standard benchmark dataset.
ieeexplore.ieee.org
Showing the best result for this search. See all results