Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2210.14598 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911842363768832 |
|---|---|
| author | Magris, Martin Shabani, Mostafa Iosifidis, Alexandros |
| author_facet | Magris, Martin Shabani, Mostafa Iosifidis, Alexandros |
| contents | We propose an optimization algorithm for Variational Inference (VI) in complex models. Our approach relies on natural gradient updates where the variational space is a Riemann manifold. We develop an efficient algorithm for Gaussian Variational Inference whose updates satisfy the positive definite constraint on the variational covariance matrix. Our Manifold Gaussian Variational Bayes on the Precision matrix (MGVBP) solution provides simple update rules, is straightforward to implement, and the use of the precision matrix parametrization has a significant computational advantage. Due to its black-box nature, MGVBP stands as a ready-to-use solution for VI in complex models. Over five datasets, we empirically validate our feasible approach on different statistical and econometric models, discussing its performance with respect to baseline methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2210_14598 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Manifold Gaussian Variational Bayes on the Precision Matrix Magris, Martin Shabani, Mostafa Iosifidis, Alexandros Machine Learning We propose an optimization algorithm for Variational Inference (VI) in complex models. Our approach relies on natural gradient updates where the variational space is a Riemann manifold. We develop an efficient algorithm for Gaussian Variational Inference whose updates satisfy the positive definite constraint on the variational covariance matrix. Our Manifold Gaussian Variational Bayes on the Precision matrix (MGVBP) solution provides simple update rules, is straightforward to implement, and the use of the precision matrix parametrization has a significant computational advantage. Due to its black-box nature, MGVBP stands as a ready-to-use solution for VI in complex models. Over five datasets, we empirically validate our feasible approach on different statistical and econometric models, discussing its performance with respect to baseline methods. |
| title | Manifold Gaussian Variational Bayes on the Precision Matrix |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2210.14598 |