Deep kernel learning for integral measurements
arXiv:1909.01844
Abstract
Deep kernel learning refers to a Gaussian process that incorporates neural networks to improve the modelling of complex functions. We present a method that makes this approach feasible for problems where the data consists of line integral measurements of the target function. The performance is illustrated on computed tomography reconstruction examples.
References in corpus (7)
- Deep Neural Networks as Gaussian Processes
- Gaussian Process Kernels for Pattern Discovery and Extrapolation
- Bayesian Optimization with Gradients
- Stochastic Variational Deep Kernel Learning
- Probabilistic modelling and reconstruction of strain
- Linearly constrained Gaussian processes
- Know Your Boundaries: Constraining Gaussian Processes by Variational Harmonic Features