BOREx: Bayesian-Optimization--Based Refinement of Saliency Map for Image- and Video-Classification Models
This addresses the challenge of generating better visual explanations for classification models, which is important for interpretability in computer vision, but it is incremental as it refines existing methods rather than introducing a new paradigm.
The paper tackles the problem of refining heat-map-based explanations for image- and video-classification models by proposing BOREx, a black-box method that uses Bayesian optimization to improve low-quality saliency maps, with experiments showing statistical improvements.
Explaining a classification result produced by an image- and video-classification model is one of the important but challenging issues in computer vision. Many methods have been proposed for producing heat-map--based explanations for this purpose, including ones based on the white-box approach that uses the internal information of a model (e.g., LRP, Grad-CAM, and Grad-CAM++) and ones based on the black-box approach that does not use any internal information (e.g., LIME, SHAP, and RISE). We propose a new black-box method BOREx (Bayesian Optimization for Refinement of visual model Explanation) to refine a heat map produced by any method. Our observation is that a heat-map--based explanation can be seen as a prior for an explanation method based on Bayesian optimization. Based on this observation, BOREx conducts Gaussian process regression (GPR) to estimate the saliency of each pixel in a given image starting from the one produced by another explanation method. Our experiments statistically demonstrate that the refinement by BOREx improves low-quality heat maps for image- and video-classification results.