首页|Studies from University of South Carolina Update Current Data on Networks (Distributed Learning Over a Wireless Network With Non-coherent Majority Vote Computation)
Studies from University of South Carolina Update Current Data on Networks (Distributed Learning Over a Wireless Network With Non-coherent Majority Vote Computation)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Network Daily News - Investigatorspublish new report on Networks. According to news reporting from Columbia, South Carolina, by NewsRxeditors, the research stated, “In this study, we propose an over-the-air computation (OAC) scheme tocalculate the majority vote (MV) for federated edge learning (FEEL). With the proposed approach, edgedevices (EDs) transmit the signs of local stochastic gradients, i.e., votes, by activating one of two orthogonalresources.”The news correspondents obtained a quote from the research from the University of South Carolina,“The MVs at the edge server (ES) are obtained with non-coherent detectors by exploiting the accumulationson the resources. Hence, the proposed scheme eliminates the need for channel state information (CSI)at the EDs and ES. In this study, we analyze various gradient-encoding strategies through the weightfunctions and waveform configurations over orthogonal frequency division multiplexing (OFDM). We showthat specific weight functions that enable absentee EDs (i.e., hard-coded participation with absentees(HPA)) or weighted votes (i.e., soft-coded participation (SP)) can substantially reduce the probability ofdetecting the incorrect MV. By taking path loss, power control, cell size, and fading channel into account,we prove the convergence of the distributed learning for a non-convex function for HPA.”
ColumbiaSouth CarolinaUnited StatesNorth and Central AmericaNetworksWireless NetworkWireless TechnologyUniversity of South Carolina
NETL
NSTL
