Benefits of Non-Linear Scale Parameterizations in Black Box Variational Inference through Smoothness Results and Gradient Variance Bounds.
Alexandra Maria HottiLennart Alexander Van der GotenJens LagergrenPublished in: AISTATS (2024)
Keyphrases
- level set
- black box
- variational inference
- variational methods
- bayesian inference
- black boxes
- white box
- image segmentation
- upper bound
- posterior distribution
- mixture model
- edge detection
- probabilistic model
- scale space
- test cases
- probabilistic graphical models
- latent dirichlet allocation
- gaussian process
- exponential family
- closed form
- prior information
- topic models
- exact inference
- worst case
- bayesian networks
- approximate inference
- data sets
- multiscale
- machine learning