Adjoint-aided inference of Gaussian process driven differential equations.
Paterne GahunguChristopher W. LanyonMauricio A. ÁlvarezEngineer BainomugishaMichael T. SmithRichard WilkinsonPublished in: NeurIPS (2022)
Keyphrases
- differential equations
- gaussian process
- covariance function
- expectation propagation
- gaussian processes
- hyperparameters
- bayesian inference
- gaussian process regression
- dynamical systems
- model selection
- semi supervised
- approximate inference
- bayesian framework
- latent variables
- variational bayes
- regression model
- optimal control
- partial differential equations
- bayesian networks
- prior information
- dynamic bayesian networks
- image processing
- gaussian process models
- noise level
- probabilistic inference
- incremental learning
- maximum likelihood
- least squares
- artificial neural networks
- optimal solution
- machine learning