This paper applies Sobol indices to compute the contribution of each feature, enabling the method to capture higher-order interactions between features. Existing methods typically use the loss function to measure the difference between the original output and the perturbed output. For example, in classification problems, cross-entropy is commonly applied as follows :
where represents the importance scores of the -th feature. In contrast, this paper computes based on Sobol indices.
The perturbation function is as follows :
In this paper, the function is defined similarly to existing literature but is referred to as Inpainting perturbation.