Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization

Abstract

Features in images' backgrounds can spuriously correlate with the images' classes, representing background bias. They can influence the classifier's decisions, causing shortcut learning (Clever Hans effect). The phenomenon generates deep neural networks (DNNs) that perform well on standard evaluation datasets but generalize poorly to real-world data. Layer-wise Relevance Propagation (LRP) explains DNNs' decisions. Here, we show that the optimization of LRP heatmaps can minimize the background bias influence on deep classifiers, hindering shortcut learning. By not increasing run-time computational cost, the approach is light and fast. Furthermore, it applies to virtually any classification architecture. After injecting synthetic bias in images' backgrounds, we compared our approach (dubbed ISNet) to eight state-of-the-art DNNs, quantitatively demonstrating its superior robustness to background bias. Mixed datasets are common for COVID-19 and tuberculosis classification with chest X-rays, fostering background bias. By focusing on the lungs, the ISNet reduced shortcut learning. Thus, its generalization performance on external (out-of-distribution) test databases significantly surpassed all implemented benchmark models.

Text and style improvements. Included reference to the published article (Nature Communications, https://doi.org/10.1038/s41467-023-44371-z)

References in corpus (10)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Shortcut Learning in Deep Neural Networks
• COVID-19 Image Data Collection
• Reliable Tuberculosis Detection using Chest X-ray with Deep Learning, Segmentation and Visualization
• Comparing Different Deep Learning Architectures for Classification of Chest Radiographs
• BS-Net: learning COVID-19 pneumonia severity on a large Chest X-Ray dataset
• Learning Hierarchical Attention for Weakly-supervised Chest X-Ray Abnormality Localization and Diagnosis
• A Deep Convolutional Neural Network for COVID-19 Detection Using Chest X-Rays
• BIMCV COVID-19+: a large annotated dataset of RX and CT images from COVID-19 patients
• Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization

Cited by in corpus (5)

• Abstractive Text Summarization: State of the Art, Challenges, and Improvements
• Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization
• Generalizable and Explainable Deep Learning for Medical Image Computing: An Overview
• Hybrid Quantum-inspired Resnet and Densenet for Pattern Recognition
• Faster ISNet for Background Bias Mitigation on Deep Neural Networks