Publications - RISE

CB-RISE: Improving the RISE Interpretability Method Through Convergence Detection and Blurred Perturbations
Conference paper

This paper enhances the RISE algorithm for image interpretation by introducing C-RISE, which reduces computation time, and CB-RISE, which uses blurred masks instead of black patches. These improvements lead to better heatmap quality and a speedup of about three times.

Black Box Blurred Masks Computer Vision Convergence Detection Deep Learning Interpretability RISE

The Implementation of the RISE Algorithm for the Captum Framework
Conference paper

This paper introduces an efficient implementation of the RISE method in PyTorch using the Captum library. RISE generates explanations for deep learning model predictions by randomly masking input parts. Experiments highlight the significance of the number and size of masks for achieving accurate and versatile interpretations.

Black Box Computer Vision Deep Learning Interpretability RISE

CB-RISE: Improving the RISE Interpretability Method Through Convergence Detection and Blurred PerturbationsConference paper

The Implementation of the RISE Algorithm for the Captum FrameworkConference paper

CB-RISE: Improving the RISE Interpretability Method Through Convergence Detection and Blurred Perturbations
Conference paper

The Implementation of the RISE Algorithm for the Captum Framework
Conference paper