Convolutional Gaussian Processes
This work addresses the problem of scaling Gaussian processes for image data, which is incremental as it adapts existing methods to a specific domain.
The authors tackled the challenge of applying Gaussian processes to high-dimensional inputs like images by introducing a convolutional structure, achieving improved performance on MNIST and CIFAR-10 datasets.
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.