A0456
Title: Nonparametric distributed learning of complex data
Authors: Guannan Wang - College of William & Mary (United States) [presenting]
Lily Wang - Iowa State University (United States)
Shan Yu - University of Virginia (United States)
Abstract: Nowadays, one significant challenge in many applications comes from the enormous size of the datasets collected from modern technologies. To tackle this challenge, a novel nonparametric distributed learning method is designed based on multivariate spline smoothing over a triangulation of the domain. The proposed DHL algorithm has a simple, scalable, and communication-efficient implementation scheme that can almost achieve linear speedup. In addition, rigorous theoretical support is provided for the DHL framework. The DHL linear estimators have proven to reach the same convergence rate as the global spline estimators obtained using the entire dataset. The proposed DHL method is evaluated through extensive simulation studies and analyses of real applications.
