Uncertainty Class Activation Map (U-CAM) Using Gradient Certainty Method

Understanding and explaining deep learning models is an imperative task. Towards this, we propose a method that obtains gradient-based certainty estimates also provide visual attention maps. Particularly, solve for question answering We incorporate modern probabilistic methods further improve by using the gradients these estimates. These have two-fold benefits: a) improvement in obtaining correlate better with misclassified samples b) improved maps state-of-the-art results terms of correlation human regions. The result consistent various answering. Therefore, proposed technique can be thought as tool explanations models. detailed empirical analysis task on all standard benchmarks comparison state art methods.