TY - GEN

T1 - Estimating signals over graphs via multi-kernel learning

AU - Romero, Daniel

AU - Ma, Meng

AU - Giannakis, Georgios B.

PY - 2016/8/24

Y1 - 2016/8/24

N2 - Estimating functions on graphs finds well-documented applications in machine learning and, more recently, in signal processing. Given signal values on a subset of vertices, the goal is to estimate the signal on the remaining ones. This task amounts to estimating a function (or signal) over a graph. Most existing techniques either rely on parametric signal models or require costly cross-validation. Leveraging the framework of multi-kernel learning, a data-driven nonparametric approach is developed here. Instead of a single kernel, the algorithm relies on a dictionary of candidate kernels and efficiently selects the most suitable ones by minimizing a convex criterion using a group Lasso module. Numerical tests demonstrate the superior estimation performance of the novel approach over competing alternatives.

AB - Estimating functions on graphs finds well-documented applications in machine learning and, more recently, in signal processing. Given signal values on a subset of vertices, the goal is to estimate the signal on the remaining ones. This task amounts to estimating a function (or signal) over a graph. Most existing techniques either rely on parametric signal models or require costly cross-validation. Leveraging the framework of multi-kernel learning, a data-driven nonparametric approach is developed here. Instead of a single kernel, the algorithm relies on a dictionary of candidate kernels and efficiently selects the most suitable ones by minimizing a convex criterion using a group Lasso module. Numerical tests demonstrate the superior estimation performance of the novel approach over competing alternatives.

KW - Graph kernels

KW - Kernel regression

KW - multi-kernel learning

UR - http://www.scopus.com/inward/record.url?scp=84987881624&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84987881624&partnerID=8YFLogxK

U2 - 10.1109/SSP.2016.7551714

DO - 10.1109/SSP.2016.7551714

M3 - Conference contribution

AN - SCOPUS:84987881624

T3 - IEEE Workshop on Statistical Signal Processing Proceedings

BT - 2016 19th IEEE Statistical Signal Processing Workshop, SSP 2016

PB - IEEE Computer Society

T2 - 19th IEEE Statistical Signal Processing Workshop, SSP 2016

Y2 - 25 June 2016 through 29 June 2016

ER -