A Rate-Distortion Framework for Explaining Black-Box Model Decisions
Stefan Kolek, Đức Anh Nguyễn, Ron Levie, Joan Bruna, Gitta Kutyniok · Lecture notes in computer science · 2022
Abstract We present theRate-Distortion Explanation(RDE) framework, a mathematically well-founded method for explaining black-box model decisions. The framework is based on perturbations of the target input signal and applies to any differentiable pre-trained model such as neural networks. Our experiments demonstrate the framework’s adaptability to diverse data modalities, particularly images, audio, and physical simulations of urban environments.