Differentiable Compositional Kernel Learning for Gaussian Processes

Abstract

The generalization properties of Gaussian processes depend heavily on the choice of kernel, and this choice remains a dark art. We present the Neural Kernel Network (NKN), a flexible family of kernels represented by a neural network. The NKN architecture is based on the composition rules for kernels, so that each unit of the network corresponds to a valid kernel. It can compactly approximate compositional kernel structures such as those used by the Automatic Statistician (Lloyd et al., 2014), but because the architecture is differentiable, it is end-to-end trainable with gradient-based optimization. We show that the NKN is universal for the class of stationary kernels. Empirically we demonstrate pattern discovery and extrapolation abilities of NKN on several tasks that depend crucially on identifying the underlying structure, including time series and texture extrapolation, as well as Bayesian optimization.

arXiv commentICML 2018; update proof

References in corpus (10)

• Practical Bayesian Optimization of Machine Learning Algorithms
• A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Structure Discovery in Nonparametric Regression through Compositional Kernel Search
• Gaussian Process Kernels for Pattern Discovery and Extrapolation
• Additive Gaussian Processes
• Exploring Large Feature Spaces with Hierarchical Multiple Kernel Learning
• Automatic Construction and Natural-Language Description of Nonparametric Regression Models
• Non-Stationary Spectral Kernels
• Generalized Spectral Kernels

Cited by in corpus (10)

• Bayesian optimization of a free-electron laser
• Deeper Connections between Neural Networks and Gaussian Processes Speed-up Active Learning
• The Gaussian Process Autoregressive Regression Model (GPAR)
• Gaussian Process Modelling for Improved Resolution in Faraday Depth Reconstruction
• Marginalised Gaussian Processes with Nested Sampling
• Sparse Spectrum Warped Input Measures for Nonstationary Kernel Learning
• Conditional Deep Gaussian Processes: multi-fidelity kernel learning
• Gaussian Processes with Skewed Laplace Spectral Mixture Kernels for Long-term Forecasting
• Kernel Identification Through Transformers
• Conditional Deep Gaussian Processes: empirical Bayes hyperdata learning