Wasserstein Gradient Flow over Variational Parameter Space for Variational Inference.
Dai Hai NguyenTetsuya SakuraiHiroshi MamitsukaPublished in: CoRR (2023)
Keyphrases
- parameter space
- variational inference
- variational methods
- curve evolution
- level set
- variational formulation
- active contour model
- active contours
- bayesian inference
- energy functional
- energy minimization
- high dimensional
- optic flow
- shape prior
- approximate inference
- level set method
- object boundaries
- hyperparameters
- optical flow
- image segmentation
- posterior distribution
- partial differential equations
- mixture model
- probabilistic model
- exponential family
- latent dirichlet allocation
- topic models
- gaussian process
- search space
- evolution equation
- feature selection
- closed form