Pseudo-Nonlinear Data Augmentation: A Constrained Energy Minimization Viewpoint
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917224059502592 |
|---|---|
| author | Hu, Pingbang Sugiyama, Mahito |
| author_facet | Hu, Pingbang Sugiyama, Mahito |
| contents | We propose a simple yet novel data augmentation method for general data modalities based on energy-based modeling and principles from information geometry. Unlike most existing learning-based data augmentation methods, which rely on learning latent representations with generative models, our proposed framework enables an intuitive construction of a geometrically aware latent space that represents the structure of the data itself, supporting efficient and explicit encoding and decoding procedures. We then present and discuss how to design latent spaces that will subsequently control the augmentation with the proposed algorithm. Empirical results demonstrate that our data augmentation method achieves competitive performance in downstream tasks compared to other baselines, while offering fine-grained controllability that is lacking in the existing literature. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_00718 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Pseudo-Nonlinear Data Augmentation: A Constrained Energy Minimization Viewpoint Hu, Pingbang Sugiyama, Mahito Machine Learning We propose a simple yet novel data augmentation method for general data modalities based on energy-based modeling and principles from information geometry. Unlike most existing learning-based data augmentation methods, which rely on learning latent representations with generative models, our proposed framework enables an intuitive construction of a geometrically aware latent space that represents the structure of the data itself, supporting efficient and explicit encoding and decoding procedures. We then present and discuss how to design latent spaces that will subsequently control the augmentation with the proposed algorithm. Empirical results demonstrate that our data augmentation method achieves competitive performance in downstream tasks compared to other baselines, while offering fine-grained controllability that is lacking in the existing literature. |
| title | Pseudo-Nonlinear Data Augmentation: A Constrained Energy Minimization Viewpoint |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2410.00718 |