This looks more like that: Enhancing Self-Explaining Models by Prototypical Relevance Propagation

TL;DR

This paper introduces Prototypical Relevance Propagation (PRP), a new explanation method that improves the precision of self-explaining models like ProtoPNet, and proposes clustering strategies to reduce artifact learning.

Contribution

The paper presents PRP, a novel explanation technique that enhances explanation accuracy and introduces clustering methods to mitigate artifact learning in self-explaining models.

Findings

PRP provides more spatially precise explanations than existing methods.

Clustering with PRP explanations effectively segregates artifact images.

The approach improves model interpretability and artifact detection.

Abstract

Current machine learning models have shown high efficiency in solving a wide variety of real-world problems. However, their black box character poses a major challenge for the understanding and traceability of the underlying decision-making strategies. As a remedy, many post-hoc explanation and self-explanatory methods have been developed to interpret the models' behavior. These methods, in addition, enable the identification of artifacts that can be learned by the model as class-relevant features. In this work, we provide a detailed case study of the self-explaining network, ProtoPNet, in the presence of a spectrum of artifacts. Accordingly, we identify the main drawbacks of ProtoPNet, especially, its coarse and spatially imprecise explanations. We address these limitations by introducing Prototypical Relevance Propagation (PRP), a novel method for generating more precise model-aware explanations. Furthermore, in order to obtain a clean dataset, we propose to use multi-view clustering strategies for segregating the artifact images using the PRP explanations, thereby suppressing the potential artifact learning in the models.