Are CNN Predictions based on Reasonable Evidence?

Sarah Adel Bargal, Andrea Zunino, Vitali Petsiuk, Jianming Zhang, Kate Saenko, Vittorio Murino, Stan Sclaroff; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019, pp. 67-70

We propose Guided Zoom, an approach that utilizes spatial grounding to make more informed predictions. It does so by making sure the model has "the right reasons" for a prediction, being defined as reasons that are coherent with those used to make similar correct decisions at training time. The reason/evidence upon which a deep neural network makes a prediction is defined to be the spatial grounding, in the pixel space, for a specific class conditional probability in the model output. Guided Zoom question show reasonable the evidence used to make a prediction is. We show that Guided Zoom results in the refinement of a model's classification accuracy on two fine-grained classification datasets.