Research
Measuring the Sensitivity of Classification Models with the Error Sensitivity Profile
arXiv:2604.25765v1 Announce Type: new Abstract: The quality of training data is critical to the performance of machine learning models. In this paper, the Error Sensitivity Profile (ESP) is proposed.
arXiv:2604.25765v1 Announce Type: new Abstract: The quality of training data is critical to the performance of machine learning models. In this paper, the Error Sensitivity Profile (ESP) is proposed. It quantifies the sensitivity of model performance to errors in a single feature or in multiple features. By leveraging ESP, data-cleaning efforts can be prioritized based on error types and features most likely to affect model performance. To support the computation of this metric, an integrated suite of tools, called irty, is created. We conduct an extensive experimental study on two widely used datasets using 14 classification models, revealing that performance degradation is not always predictable from simple correlations with the target variable.
Related
- Exploring the Impact of Dataset Statistical Effect Size on Model Performance and Data Sample Size Sufficiency
- Toward Efficient Influence Function: Dropout as a Compression Tool
- Structured Exploration and Exploitation of Label Functions for Automated Data Annotation
- High-dimensional Semi-supervised Classification via the Fermat Distance
Source: arXiv cs.LG | 2026-04-29