Search results with tag "To end deep learning architecture for graph"

An End-to-End Deep Learning Architecture for Graph ...

by graph Fourier transform. This transformation involves expensive multiplications with the eigenvector matrix of the graph Laplacian. To reduce the computation burden, (Def-ferrard, Bresson, and Vandergheynst 2016) parameterized the spectral filters as Chebyshev polynomials of eigenvalues, and achieved efficient and localized filters.

  Architecture, Learning, Deep, Graph, Fourier, Spectral, Chebyshev, To end deep learning architecture for graph