Automated Processing of eXplainable Artificial Intelligence Outputs in Deep Learning Models for Fault Diagnostics of Large Infrastructures

2025-03-19

AI Summary

Deep learning models for large-scale infrastructure health diagnostics often rely on non-causal shortcuts—such as ID tags—yielding “correct but untrustworthy” predictions. To address this, we propose the first automated framework that tightly integrates posterior explainable AI (XAI) interpretation (via Grad-CAM) with deep semi-supervised anomaly detection (Deep SSAD), enabling end-to-end identification and explanation of anomalies. Unlike conventional approaches that optimize solely for classification accuracy or explanation fidelity, our method precisely localizes model failures driven by spurious correlations. Evaluated on drone-captured insulator housing defect diagnosis, the framework improves average classification accuracy by 8% across two fault types, reduces manual verification effort to 15%, and achieves significantly higher F₁ scores than state-of-the-art fidelity-based methods.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Semi-Supervised Learning
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Deep Learning (DL) models processing images to recognize the health state of large infrastructure components can exhibit biases and rely on non-causal shortcuts. eXplainable Artificial Intelligence (XAI) can address these issues but manually analyzing explanations generated by XAI techniques is time-consuming and prone to errors. This work proposes a novel framework that combines post-hoc explanations with semi-supervised learning to automatically identify anomalous explanations that deviate from those of correctly classified images and may therefore indicate model abnormal behaviors. This significantly reduces the workload for maintenance decision-makers, who only need to manually reclassify images flagged as having anomalous explanations. The proposed framework is applied to drone-collected images of insulator shells for power grid infrastructure monitoring, considering two different Convolutional Neural Networks (CNNs), GradCAM explanations and Deep Semi-Supervised Anomaly Detection. The average classification accuracy on two faulty classes is improved by 8% and maintenance operators are required to manually reclassify only 15% of the images. We compare the proposed framework with a state-of-the-art approach based on the faithfulness metric: the experimental results obtained demonstrate that the proposed framework consistently achieves F_1 scores larger than those of the faithfulness-based approach. Additionally, the proposed framework successfully identifies correct classifications that result from non-causal shortcuts, such as the presence of ID tags printed on insulator shells.

Problem

• Automates XAI output analysis for DL fault diagnostics
• Reduces manual workload by flagging anomalous explanations
• Improves classification accuracy and detects non-causal shortcuts

Innovation

• Combines XAI with semi-supervised learning
• Automates anomaly detection in model explanations
• Improves accuracy and reduces manual reclassification