1: \begin{abstract}

2: When facing graph signal processing tasks, the workhorse assumption is that the graph describing the support of the signals is known.

3: However, in many relevant applications \emph{the available graph suffers from observation errors and perturbations}.

4: As a result, any method relying on the graph topology may yield suboptimal results if those imperfections are ignored.  

5: 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.

6: 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.

7: %As a result, on top of estimating the desired GF at hand, a modified (true) graph is obtained as a byproduct.

8: As a result, on top of learning the desired GF, an estimate of the graph is obtained as a byproduct.

9: 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.

10: The second part of the paper i) generalizes the design to a robust setup where several GFs %(all defined over the same graph) 

11: are jointly estimated, and ii) introduces an alternative algorithmic implementation that reduces the computational complexity. %when dealing with large graphs.

12: 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. 

13: \end{abstract}

14: