Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2411.03097 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911085106298880 |
|---|---|
| author | Orme, Ella S. C. Evangelou, Marina Paquet, Ulrich |
| author_facet | Orme, Ella S. C. Evangelou, Marina Paquet, Ulrich |
| contents | Multi-view data from the same source often exhibit correlation. This is mirrored in correlation between the latent spaces of separate variational autoencoders (VAEs) trained on each data-view. A multi-view VAE approach is proposed that incorporates a joint prior with a non-zero correlation structure between the latent spaces of the VAEs. By enforcing such correlation structure, more strongly correlated latent spaces are uncovered. Using conditional distributions to move between these latent spaces, missing views can be imputed and used for downstream analysis. Learning this correlation structure involves maintaining validity of the prior distribution, as well as a successful parameterization that allows end-to-end learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_03097 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Correlating Variational Autoencoders Natively For Multi-View Imputation Orme, Ella S. C. Evangelou, Marina Paquet, Ulrich Machine Learning Multi-view data from the same source often exhibit correlation. This is mirrored in correlation between the latent spaces of separate variational autoencoders (VAEs) trained on each data-view. A multi-view VAE approach is proposed that incorporates a joint prior with a non-zero correlation structure between the latent spaces of the VAEs. By enforcing such correlation structure, more strongly correlated latent spaces are uncovered. Using conditional distributions to move between these latent spaces, missing views can be imputed and used for downstream analysis. Learning this correlation structure involves maintaining validity of the prior distribution, as well as a successful parameterization that allows end-to-end learning. |
| title | Correlating Variational Autoencoders Natively For Multi-View Imputation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2411.03097 |