Explainable Supervised Domain Adaptation

TL;DR

This paper introduces XSDA-Net, an explainable supervised domain adaptation framework that uses case-based reasoning to clarify predictions by highlighting similar regions in source and target images, enhancing interpretability.

Contribution

The paper presents a novel explainable domain adaptation method integrating case-based reasoning, addressing the lack of interpretability in existing techniques.

Findings

Demonstrates the effectiveness of XSDA-Net on datasets with part-based explainability

Provides insights into the regions influencing predictions in domain adaptation

Enhances trust and transparency in deep learning models for domain adaptation

Abstract

Domain adaptation techniques have contributed to the success of deep learning. Leveraging knowledge from an auxiliary source domain for learning in labeled data-scarce target domain is fundamental to domain adaptation. While these techniques result in increasing accuracy, the adaptation process, particularly the knowledge leveraged from the source domain, remains unclear. This paper proposes an explainable by design supervised domain adaptation framework - XSDA-Net. We integrate a case-based reasoning mechanism into the XSDA-Net to explain the prediction of a test instance in terms of similar-looking regions in the source and target train images. We empirically demonstrate the utility of the proposed framework by curating the domain adaptation settings on datasets popularly known to exhibit part-based explainability.