English

AS-XAI: Self-supervised Automatic Semantic Interpretation for CNN

Computer Vision and Pattern Recognition 2023-12-27 v1 Artificial Intelligence Human-Computer Interaction Information Retrieval Machine Learning

Abstract

Explainable artificial intelligence (XAI) aims to develop transparent explanatory approaches for "black-box" deep learning models. However,it remains difficult for existing methods to achieve the trade-off of the three key criteria in interpretability, namely, reliability, causality, and usability, which hinder their practical applications. In this paper, we propose a self-supervised automatic semantic interpretable explainable artificial intelligence (AS-XAI) framework, which utilizes transparent orthogonal embedding semantic extraction spaces and row-centered principal component analysis (PCA) for global semantic interpretation of model decisions in the absence of human interference, without additional computational costs. In addition, the invariance of filter feature high-rank decomposition is used to evaluate model sensitivity to different semantic concepts. Extensive experiments demonstrate that robust and orthogonal semantic spaces can be automatically extracted by AS-XAI, providing more effective global interpretability for convolutional neural networks (CNNs) and generating human-comprehensible explanations. The proposed approach offers broad fine-grained extensible practical applications, including shared semantic interpretation under out-of-distribution (OOD) categories, auxiliary explanations for species that are challenging to distinguish, and classification explanations from various perspectives.

Keywords

Cite

@article{arxiv.2312.14935,
  title  = {AS-XAI: Self-supervised Automatic Semantic Interpretation for CNN},
  author = {Changqi Sun and Hao Xu and Yuntian Chen and Dongxiao Zhang},
  journal= {arXiv preprint arXiv:2312.14935},
  year   = {2023}
}
R2 v1 2026-06-28T14:00:14.896Z