Generating Part-based Global Explnations via Correspondence
Published in IJCAI 2024 Workshop on Explainable Artificial Intelligence (XAI), 2024
Deep learning models are notoriously opaque. Existing explanation methods often focus on localized visual explanations for individual images. Concept-based explanations, while offering global insights, require extensive annotations, incurring significant labeling cost. We propose an approach that leverages user-defined part labels from a limited set of images and efficiently transfers them to a larger dataset. This enables the generation of global symbolic explanations by aggregating part-based local explanations, ultimately providing human-understandable explanations for model decisions on a large scale.
Recommended citation: Kunal Rathore, (2024). "Generating Part-based Global Explnations via Correspondence." IJCAI 1. 1(1). https://drive.google.com/file/d/1_1LySYzjUUow22PV4M9FScRstdFIzk89/view