Batch and match: black-box variational inference with a score-based divergence
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913388224839680 |
|---|---|
| author | Cai, Diana Modi, Chirag Pillaud-Vivien, Loucas Margossian, Charles C. Gower, Robert M. Blei, David M. Saul, Lawrence K. |
| author_facet | Cai, Diana Modi, Chirag Pillaud-Vivien, Loucas Margossian, Charles C. Gower, Robert M. Blei, David M. Saul, Lawrence K. |
| contents | Most leading implementations of black-box variational inference (BBVI) are based on optimizing a stochastic evidence lower bound (ELBO). But such approaches to BBVI often converge slowly due to the high variance of their gradient estimates and their sensitivity to hyperparameters. In this work, we propose batch and match (BaM), an alternative approach to BBVI based on a score-based divergence. Notably, this score-based divergence can be optimized by a closed-form proximal update for Gaussian variational families with full covariance matrices. We analyze the convergence of BaM when the target distribution is Gaussian, and we prove that in the limit of infinite batch size the variational parameter updates converge exponentially quickly to the target mean and covariance. We also evaluate the performance of BaM on Gaussian and non-Gaussian target distributions that arise from posterior inference in hierarchical and deep generative models. In these experiments, we find that BaM typically converges in fewer (and sometimes significantly fewer) gradient evaluations than leading implementations of BBVI based on ELBO maximization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_14758 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Batch and match: black-box variational inference with a score-based divergence Cai, Diana Modi, Chirag Pillaud-Vivien, Loucas Margossian, Charles C. Gower, Robert M. Blei, David M. Saul, Lawrence K. Machine Learning Artificial Intelligence Computation Most leading implementations of black-box variational inference (BBVI) are based on optimizing a stochastic evidence lower bound (ELBO). But such approaches to BBVI often converge slowly due to the high variance of their gradient estimates and their sensitivity to hyperparameters. In this work, we propose batch and match (BaM), an alternative approach to BBVI based on a score-based divergence. Notably, this score-based divergence can be optimized by a closed-form proximal update for Gaussian variational families with full covariance matrices. We analyze the convergence of BaM when the target distribution is Gaussian, and we prove that in the limit of infinite batch size the variational parameter updates converge exponentially quickly to the target mean and covariance. We also evaluate the performance of BaM on Gaussian and non-Gaussian target distributions that arise from posterior inference in hierarchical and deep generative models. In these experiments, we find that BaM typically converges in fewer (and sometimes significantly fewer) gradient evaluations than leading implementations of BBVI based on ELBO maximization. |
| title | Batch and match: black-box variational inference with a score-based divergence |
| topic | Machine Learning Artificial Intelligence Computation |
| url | https://arxiv.org/abs/2402.14758 |