Node varying regularization for graph signals

Node varying regularization for graph signals

Yang, M. (TU Delft Electrical Engineering, Mathematics and Computer Science)
Coutino, Mario (TU Delft Signal Processing Systems)
Isufi, E. (TU Delft Multimedia Computing)
Leus, G.J.T. (TU Delft Signal Processing Systems)

Electrical Engineering, Mathematics and Computer Science

2020

While regularization on graphs has been successful for signal reconstruction, strategies for controlling the bias-variance trade-off of such methods have not been completely explored. In this work, we put forth a node varying regularizer for graph signal reconstruction and develop a minmax approach to design the vector of regularization parameters. The proposed design only requires as prior information an upper bound on the underlying signal energy; a reasonable assumption in practice. With such formulation, an iterative method is introduced to obtain a solution meeting global equilibrium. The approach is numerically efficient and has convergence guarantees. Numerical simulations using real data support the proposed design scheme.

Bias-variance trade-off
Graph regularization
Graph signal denoising
Graph signal processing
Minmax problems

http://resolver.tudelft.nl/uuid:2350d6dc-a45b-444a-a6ea-0f22bc3eea69

https://doi.org/10.23919/Eusipco47968.2020.9287807

Eurasip, Amsterdam (Netherlands)

2021-08-29

978-9-0827-9705-3

28th European Signal Processing Conference, EUSIPCO 2020 - Proceedings

EUSIPCO 2020, 2021-01-18 → 2021-01-22, Amsterdam, Netherlands

European Signal Processing Conference, 2219-5491, 2021-January

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. Date change due to COVID-19 (former date August 24-28 2020)

Institutional Repository

conference paper

© 2020 M. Yang, Mario Coutino, E. Isufi, G.J.T. Leus