Improving the Gaussian Process Sparse Spectrum Approximation by Representing Uncertainty in Frequency Inputs
arXiv:1503.02424
Abstract
Standard sparse pseudo-input approximations to the Gaussian process (GP) cannot handle complex functions well. Sparse spectrum alternatives attempt to answer this but are known to over-fit. We suggest the use of variational inference for the sparse spectrum approximation to avoid both issues. We model the covariance function with a finite Fourier series approximation and treat it as a random variable. The random covariance function has a posterior, on which a variational distribution is placed. The variational distribution transforms the random covariance function to fit the data. We study the properties of our approximate inference, compare it to alternative ones, and extend it to the distributed and stochastic domains. Our approximation captures complex functions better than standard approaches and avoids over-fitting.
13 pages, 3 figures
References in corpus (4)
- A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
- Gaussian Processes for Big Data
- Structure Discovery in Nonparametric Regression through Compositional Kernel Search
- Variational inference for sparse spectrum Gaussian process regression
Cited by in corpus (11)
- Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference
- When Gaussian Process Meets Big Data: A Review of Scalable GPs
- Dropout as a Bayesian Approximation: Appendix
- Practical Hilbert space approximate Bayesian Gaussian processes for probabilistic programming
- Variational Implicit Processes
- Stochastic Variational Inference for Bayesian Sparse Gaussian Process Regression
- Real-Time Regression with Dividing Local Gaussian Processes
- Scalable Training of Inference Networks for Gaussian-Process Models
- Inter-domain Deep Gaussian Processes
- Marginalising over Stationary Kernels with Bayesian Quadrature
- Spectrum Gaussian Processes Based On Tunable Basis Functions