A Robust Alternative for Graph Convolutional Neural Networks via Graph Neighborhood Filters

Tenorio, Víctor M.; Rey, Samuel; Gama, Fernando; Segarra, Santiago; Marqués, Antonio G.

doi:10.1109/ieeeconf53345.2021.9723282

Cited by 6 publications

(1 citation statement)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the presence of perturbations has also been considered in non-linear GSP tasks. An alternative definition of GFs robust to perturbations is proposed in [29], and the transferability of GFs when employed in graph neural networks is studied in [30]- [32].…”

Section: Introductionmentioning

confidence: 99%

Robust Graph Filter Identification and Graph Denoising from Signal Observations

Rey¹,

Tenorio²,

Marqués³

2022

Preprint

View full text Add to dashboard Cite

When facing graph signal processing tasks, the workhorse assumption is that the graph describing the support of the signals is known. However, in many relevant applications the available graph suffers from observation errors and perturbations. As a result, any method relying on the graph topology may yield suboptimal results if those imperfections are ignored. Motivated by this, we propose a novel approach for handling perturbations on the links of the graph and apply it to the problem of robust graph filter (GF) identification from input-output observations. Different from existing works, we formulate a non-convex optimization problem that operates in the vertex domain and jointly performs GF identification and graph denoising. As a result, on top of learning the desired GF, an estimate of the graph is obtained as a byproduct. To handle the resulting bi-convex problem, we design an algorithm that blends techniques from alternating optimization and majorization minimization, showing its convergence to a stationary point. The second part of the paper i) generalizes the design to a robust setup where several GFs are jointly estimated, and ii) introduces an alternative algorithmic implementation that reduces the computational complexity. Finally, the detrimental influence of the perturbations and the benefits resulting from the robust approach are numerically analyzed over synthetic and real-world datasets, comparing them with other state-of-the-art alternatives.

show abstract

Section: Introductionmentioning

confidence: 99%