Uncovering Memorization Effect in the Presence of Spurious Correlations
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916780034752512 |
|---|---|
| author | You, Chenyu Dai, Haocheng Min, Yifei Sekhon, Jasjeet S. Joshi, Sarang Duncan, James S. |
| author_facet | You, Chenyu Dai, Haocheng Min, Yifei Sekhon, Jasjeet S. Joshi, Sarang Duncan, James S. |
| contents | Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on atypical examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we hypothesize that spurious memorization, concentrated within a small subset of neurons, plays a key role in driving imbalanced group performance. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_00961 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Uncovering Memorization Effect in the Presence of Spurious Correlations You, Chenyu Dai, Haocheng Min, Yifei Sekhon, Jasjeet S. Joshi, Sarang Duncan, James S. Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Image and Video Processing Machine learning models often rely on simple spurious features -- patterns in training data that correlate with targets but are not causally related to them, like image backgrounds in foreground classification. This reliance typically leads to imbalanced test performance across minority and majority groups. In this work, we take a closer look at the fundamental cause of such imbalanced performance through the lens of memorization, which refers to the ability to predict accurately on atypical examples (minority groups) in the training set but failing in achieving the same accuracy in the testing set. This paper systematically shows the ubiquitous existence of spurious features in a small set of neurons within the network, providing the first-ever evidence that memorization may contribute to imbalanced group performance. Through three experimental sources of converging empirical evidence, we find the property of a small subset of neurons or channels in memorizing minority group information. Inspired by these findings, we hypothesize that spurious memorization, concentrated within a small subset of neurons, plays a key role in driving imbalanced group performance. To further substantiate this hypothesis, we show that eliminating these unnecessary spurious memorization patterns via a novel framework during training can significantly affect the model performance on minority groups. Our experimental results across various architectures and benchmarks offer new insights on how neural networks encode core and spurious knowledge, laying the groundwork for future research in demystifying robustness to spurious correlation. |
| title | Uncovering Memorization Effect in the Presence of Spurious Correlations |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Image and Video Processing |
| url | https://arxiv.org/abs/2501.00961 |