Transformation Models for Flexible Posteriors in Variational Bayes
arXiv:2106.00528
Abstract
The main challenge in Bayesian models is to determine the posterior for the model parameters. Already, in models with only one or few parameters, the analytical posterior can only be determined in special settings. In Bayesian neural networks, variational inference is widely used to approximate difficult-to-compute posteriors by variational distributions. Usually, Gaussians are used as variational distributions (Gaussian-VI) which limits the quality of the approximation due to their limited flexibility. Transformation models on the other hand are flexible enough to fit any distribution. Here we present transformation model-based variational inference (TM-VI) and demonstrate that it allows to accurately approximate complex posteriors in models with one parameter and also works in a mean-field fashion for multi-parameter models like neural networks.
arXiv comment5 pages, 4 figures
References in corpus (10)
- NICE: Non-linear Independent Components Estimation
- Variational Inference with Normalizing Flows
- Weight Uncertainty in Neural Networks
- Normalizing Flows: An Introduction and Review of Current Methods
- Multiplicative Normalizing Flows for Variational Bayesian Neural Networks
- Most Likely Transformations
- Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors
- Liberty or Depth: Deep Bayesian Neural Nets Do Not Need Complex Weight Posterior Approximations
- Deep Conditional Transformation Models
- Robust normalizing flows using Bernstein-type polynomials