An Analytical Model for Overparameterized Learning Under Class Imbalance
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929747177504768 |
|---|---|
| author | Mor, Eliav Carmon, Yair |
| author_facet | Mor, Eliav Carmon, Yair |
| contents | We study class-imbalanced linear classification in a high-dimensional Gaussian mixture model. We develop a tight, closed form approximation for the test error of several practical learning methods, including logit adjustment and class dependent temperature. Our approximation allows us to analytically tune and compare these methods, highlighting how and when they overcome the pitfalls of standard cross-entropy minimization. We test our theoretical findings on simulated data and imbalanced CIFAR10, MNIST and FashionMNIST datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_05289 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | An Analytical Model for Overparameterized Learning Under Class Imbalance Mor, Eliav Carmon, Yair Machine Learning We study class-imbalanced linear classification in a high-dimensional Gaussian mixture model. We develop a tight, closed form approximation for the test error of several practical learning methods, including logit adjustment and class dependent temperature. Our approximation allows us to analytically tune and compare these methods, highlighting how and when they overcome the pitfalls of standard cross-entropy minimization. We test our theoretical findings on simulated data and imbalanced CIFAR10, MNIST and FashionMNIST datasets. |
| title | An Analytical Model for Overparameterized Learning Under Class Imbalance |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2503.05289 |