SepVAE: a contrastive VAE to separate pathological patterns from healthy ones
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917632076152832 |
|---|---|
| author | Louiset, Robin Duchesnay, Edouard Grigis, Antoine Dufumier, Benoit Gori, Pietro |
| author_facet | Louiset, Robin Duchesnay, Edouard Grigis, Antoine Dufumier, Benoit Gori, Pietro |
| contents | Contrastive Analysis VAE (CA-VAEs) is a family of Variational auto-encoders (VAEs) that aims at separating the common factors of variation between a background dataset (BG) (i.e., healthy subjects) and a target dataset (TG) (i.e., patients) from the ones that only exist in the target dataset. To do so, these methods separate the latent space into a set of salient features (i.e., proper to the target dataset) and a set of common features (i.e., exist in both datasets). Currently, all models fail to prevent the sharing of information between latent spaces effectively and to capture all salient factors of variation. To this end, we introduce two crucial regularization losses: a disentangling term between common and salient representations and a classification term between background and target samples in the salient space. We show a better performance than previous CA-VAEs methods on three medical applications and a natural images dataset (CelebA). Code and datasets are available on GitHub https://github.com/neurospin-projects/2023_rlouiset_sepvae. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_06206 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | SepVAE: a contrastive VAE to separate pathological patterns from healthy ones Louiset, Robin Duchesnay, Edouard Grigis, Antoine Dufumier, Benoit Gori, Pietro Computer Vision and Pattern Recognition Machine Learning Contrastive Analysis VAE (CA-VAEs) is a family of Variational auto-encoders (VAEs) that aims at separating the common factors of variation between a background dataset (BG) (i.e., healthy subjects) and a target dataset (TG) (i.e., patients) from the ones that only exist in the target dataset. To do so, these methods separate the latent space into a set of salient features (i.e., proper to the target dataset) and a set of common features (i.e., exist in both datasets). Currently, all models fail to prevent the sharing of information between latent spaces effectively and to capture all salient factors of variation. To this end, we introduce two crucial regularization losses: a disentangling term between common and salient representations and a classification term between background and target samples in the salient space. We show a better performance than previous CA-VAEs methods on three medical applications and a natural images dataset (CelebA). Code and datasets are available on GitHub https://github.com/neurospin-projects/2023_rlouiset_sepvae. |
| title | SepVAE: a contrastive VAE to separate pathological patterns from healthy ones |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2307.06206 |