Improving Explainability of Disentangled Representations using Multipath-Attribution Mappings

TL;DR

This paper introduces a framework that enhances explainability in AI models by using disentangled representations and multi-path attribution mappings, aiding causal analysis and robustness in safety-critical medical applications.

Contribution

The proposed method combines interpretable disentangled representations with multi-path attribution to improve explanation quality and causal insight in AI models.

Findings

Effective on synthetic and medical datasets

Enables causal relation investigation

Improves model robustness and shortcut detection

Abstract

Explainable AI aims to render model behavior understandable by humans, which can be seen as an intermediate step in extracting causal relations from correlative patterns. Due to the high risk of possible fatal decisions in image-based clinical diagnostics, it is necessary to integrate explainable AI into these safety-critical systems. Current explanatory methods typically assign attribution scores to pixel regions in the input image, indicating their importance for a model's decision. However, they fall short when explaining why a visual feature is used. We propose a framework that utilizes interpretable disentangled representations for downstream-task prediction. Through visualizing the disentangled representations, we enable experts to investigate possible causation effects by leveraging their domain knowledge. Additionally, we deploy a multi-path attribution mapping for enriching and validating explanations. We demonstrate the effectiveness of our approach on a synthetic benchmark suite and two medical datasets. We show that the framework not only acts as a catalyst for causal relation extraction but also enhances model robustness by enabling shortcut detection without the need for testing under distribution shifts.