Improving the performance of Stein variational inference through extreme sparsification of physically-constrained neural network models.
Govinda Anantha PadmanabhaJan Niklas FuhgCosmin SaftaReese E. JonesNikolaos BouklasPublished in: CoRR (2024)
Keyphrases
- neural network model
- variational inference
- bayesian models
- bayesian inference
- neural network
- topic models
- probabilistic graphical models
- posterior distribution
- latent dirichlet allocation
- variational methods
- mixture model
- gaussian process
- probabilistic model
- artificial neural networks
- closed form
- exact inference
- approximate inference
- exponential family
- least squares
- factor graphs
- latent variables
- bayesian framework
- hyperparameters
- statistical models
- regression model
- model selection
- maximum likelihood
- graphical models