FreeAugment: Data Augmentation Search Across All Degrees of Freedom
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909307985985536 |
|---|---|
| author | Bekor, Tom Nayman, Niv Zelnik-Manor, Lihi |
| author_facet | Bekor, Tom Nayman, Niv Zelnik-Manor, Lihi |
| contents | Data augmentation has become an integral part of deep learning, as it is known to improve the generalization capabilities of neural networks. Since the most effective set of image transformations differs between tasks and domains, automatic data augmentation search aims to alleviate the extreme burden of manually finding the optimal image transformations. However, current methods are not able to jointly optimize all degrees of freedom: (1) the number of transformations to be applied, their (2) types, (3) order, and (4) magnitudes. Many existing methods risk picking the same transformation more than once, limit the search to two transformations only, or search for the number of transformations exhaustively or iteratively in a myopic manner. Our approach, FreeAugment, is the first to achieve global optimization of all four degrees of freedom simultaneously, using a fully differentiable method. It efficiently learns the number of transformations and a probability distribution over their permutations, inherently refraining from redundant repetition while sampling. Our experiments demonstrate that this joint learning of all degrees of freedom significantly improves performance, achieving state-of-the-art results on various natural image benchmarks and beyond across other domains. Project page at https://tombekor.github.io/FreeAugment-web |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_04820 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FreeAugment: Data Augmentation Search Across All Degrees of Freedom Bekor, Tom Nayman, Niv Zelnik-Manor, Lihi Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Optimization and Control I.2; I.4 Data augmentation has become an integral part of deep learning, as it is known to improve the generalization capabilities of neural networks. Since the most effective set of image transformations differs between tasks and domains, automatic data augmentation search aims to alleviate the extreme burden of manually finding the optimal image transformations. However, current methods are not able to jointly optimize all degrees of freedom: (1) the number of transformations to be applied, their (2) types, (3) order, and (4) magnitudes. Many existing methods risk picking the same transformation more than once, limit the search to two transformations only, or search for the number of transformations exhaustively or iteratively in a myopic manner. Our approach, FreeAugment, is the first to achieve global optimization of all four degrees of freedom simultaneously, using a fully differentiable method. It efficiently learns the number of transformations and a probability distribution over their permutations, inherently refraining from redundant repetition while sampling. Our experiments demonstrate that this joint learning of all degrees of freedom significantly improves performance, achieving state-of-the-art results on various natural image benchmarks and beyond across other domains. Project page at https://tombekor.github.io/FreeAugment-web |
| title | FreeAugment: Data Augmentation Search Across All Degrees of Freedom |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Optimization and Control I.2; I.4 |
| url | https://arxiv.org/abs/2409.04820 |