Convolutional Gaussian Processes

Abstract

We present a practical way of introducing convolutional structure into Gaussian processes, making them more suited to high-dimensional inputs like images. The main contribution of our work is the construction of an inter-domain inducing point approximation that is well-tailored to the convolutional kernel. This allows us to gain the generalisation benefit of a convolutional kernel, together with fast but accurate posterior inference. We investigate several variations of the convolutional kernel, and apply it to MNIST and CIFAR-10, which have both been known to be challenging for Gaussian processes. We also show how the marginal likelihood can be used to find an optimal weighting between convolutional and RBF kernels to further improve performance. We hope that this illustration of the usefulness of a marginal likelihood will help automate discovering architectures in larger models.

To appear in Advances in Neural Information Processing Systems 30 (NIPS 2017)

References in corpus (1)

• Consistency and robustness of kernel-based regression in convex risk minimization

Cited by in corpus (12)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Towards Explaining Anomalies: A Deep Taylor Decomposition of One-Class Models
• Deep Convolutional Networks as shallow Gaussian Processes
• On Feature Collapse and Deep Kernel Learning for Single Forward Pass Uncertainty
• Deep Gaussian Processes for Biogeophysical Parameter Retrieval and Model Inversion
• Deep Gaussian Processes with Convolutional Kernels
• Global inducing point variational posteriors for Bayesian neural networks and deep Gaussian processes
• Graph Convolutional Gaussian Processes For Link Prediction
• GPflux: A Library for Deep Gaussian Processes
• NIPS - Not Even Wrong? A Systematic Review of Empirically Complete Demonstrations of Algorithmic Effectiveness in the Machine Learning and Artificial Intelligence Literature
• A Bayesian Approach to Invariant Deep Neural Networks