MoreauGrad: Sparse and Robust Interpretation of Neural Networks via Moreau Envelope

TL;DR

MoreauGrad introduces a robust, smooth interpretation method for neural networks based on the Moreau envelope, effectively incorporating sparsity and group-sparsity priors, and demonstrating superior interpretability in vision tasks.

Contribution

It proposes MoreauGrad, a novel interpretation scheme leveraging the Moreau envelope for robustness and sparsity, with efficient computation and applicability to various regularization techniques.

Findings

MoreauGrad provides more robust explanations than standard gradient methods.

The scheme effectively incorporates sparsity and group-sparsity priors.

Empirical results show improved interpretability on vision datasets.

Abstract

Explaining the predictions of deep neural nets has been a topic of great interest in the computer vision literature. While several gradient-based interpretation schemes have been proposed to reveal the influential variables in a neural net's prediction, standard gradient-based interpretation frameworks have been commonly observed to lack robustness to input perturbations and flexibility for incorporating prior knowledge of sparsity and group-sparsity structures. In this work, we propose MoreauGrad as an interpretation scheme based on the classifier neural net's Moreau envelope. We demonstrate that MoreauGrad results in a smooth and robust interpretation of a multi-layer neural network and can be efficiently computed through first-order optimization methods. Furthermore, we show that MoreauGrad can be naturally combined with $L_1$-norm regularization techniques to output a sparse or group-sparse explanation which are prior conditions applicable to a wide range of deep learning applications. We empirically evaluate the proposed MoreauGrad scheme on standard computer vision datasets, showing the qualitative and quantitative success of the MoreauGrad approach in comparison to standard gradient-based interpretation methods.