On Signal-to-Noise Ratio Issues in Variational Inference for Deep Gaussian Processes.
Tim G. J. RudnerOscar KeyYarin GalTom RainforthPublished in: CoRR (2020)
Keyphrases
- signal to noise ratio
- gaussian processes
- gaussian process
- variational inference
- noise reduction
- hyperparameters
- regression model
- bayesian inference
- posterior distribution
- approximate inference
- bayesian framework
- model selection
- latent variables
- multi task learning
- semi supervised
- probabilistic model
- closed form
- topic models
- variational methods
- image processing
- missing data
- maximum likelihood
- edge detection
- cross validation
- non stationary
- mixture model
- probability distribution
- support vector
- clustering algorithm
- learning algorithm