EXACT: How to Train Your Accuracy
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2022
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929433512771584 |
|---|---|
| author | Karpukhin, Ivan Dereka, Stanislav Kolesnikov, Sergey |
| author_facet | Karpukhin, Ivan Dereka, Stanislav Kolesnikov, Sergey |
| contents | Classification tasks are usually evaluated in terms of accuracy. However, accuracy is discontinuous and cannot be directly optimized using gradient ascent. Popular methods minimize cross-entropy, hinge loss, or other surrogate losses, which can lead to suboptimal results. In this paper, we propose a new optimization framework by introducing stochasticity to a model's output and optimizing expected accuracy, i.e. accuracy of the stochastic model. Extensive experiments on linear models and deep image classification show that the proposed optimization method is a powerful alternative to widely used classification losses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2205_09615 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | EXACT: How to Train Your Accuracy Karpukhin, Ivan Dereka, Stanislav Kolesnikov, Sergey Machine Learning Computer Vision and Pattern Recognition Classification tasks are usually evaluated in terms of accuracy. However, accuracy is discontinuous and cannot be directly optimized using gradient ascent. Popular methods minimize cross-entropy, hinge loss, or other surrogate losses, which can lead to suboptimal results. In this paper, we propose a new optimization framework by introducing stochasticity to a model's output and optimizing expected accuracy, i.e. accuracy of the stochastic model. Extensive experiments on linear models and deep image classification show that the proposed optimization method is a powerful alternative to widely used classification losses. |
| title | EXACT: How to Train Your Accuracy |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2205.09615 |