Fast optimize-and-sample method for differentiable Galerkin approximations of multi-layered Gaussian process priors.
Muhammad F. EmzirNiki A. LoppiZheng ZhaoSyeda Sakira HassanSimo SärkkäPublished in: FUSION (2022)
Keyphrases
- multi layered
- gaussian process
- bayesian framework
- objective function
- probabilistic model
- model selection
- decision trees
- learning algorithm
- prior knowledge
- expectation propagation
- prior information
- sparse approximations
- covariance function
- gaussian processes
- image reconstruction
- sample size
- error rate
- non stationary
- semi supervised
- image segmentation