HiBug2: Efficient and Interpretable Error Slice Discovery for Comprehensive Model Debugging
This addresses the need for robust and reliable deep learning models in real-world scenarios by automating error slice discovery and repair, though it appears incremental as it builds on prior error slice approaches.
The paper tackles the problem of systematic failures in deep learning models on specific data subsets (error slices) by introducing HiBug2, an automated framework for error slice discovery and model repair. The results show that HiBug2 improves the coherence and precision of identified error slices and significantly enhances model repair capabilities across multiple domains like image classification, pose estimation, and object detection.
Despite the significant success of deep learning models in computer vision, they often exhibit systematic failures on specific data subsets, known as error slices. Identifying and mitigating these error slices is crucial to enhancing model robustness and reliability in real-world scenarios. In this paper, we introduce HiBug2, an automated framework for error slice discovery and model repair. HiBug2 first generates task-specific visual attributes to highlight instances prone to errors through an interpretable and structured process. It then employs an efficient slice enumeration algorithm to systematically identify error slices, overcoming the combinatorial challenges that arise during slice exploration. Additionally, HiBug2 extends its capabilities by predicting error slices beyond the validation set, addressing a key limitation of prior approaches. Extensive experiments across multiple domains, including image classification, pose estimation, and object detection - show that HiBug2 not only improves the coherence and precision of identified error slices but also significantly enhances the model repair capabilities.