Scalable Gaussian Processes with Billions of Inducing Inputs via Tensor Train Decomposition
This topic contains 0 replies, has 1 voice, and was last updated by arXiv 1 year ago.

Scalable Gaussian Processes with Billions of Inducing Inputs via Tensor Train Decomposition
We propose a method (TTGP) for approximate inference in Gaussian Process (GP) models. We build on previous scalable GP research including stochastic variational inference based on inducing inputs, kernel interpolation, and structure exploiting algebra. The key idea of our method is to use Tensor Train decomposition for variational parameters, which allows us to train GPs with billions of inducing inputs and achieve stateoftheart results on several benchmarks. Further, our approach allows for training kernels based on deep neural networks without any modifications to the underlying GP model. A neural network learns a multidimensional embedding for the data, which is used by the GP to make the final prediction. We train GP and neural network parameters endtoend without pretraining, through maximization of GP marginal likelihood. We show the efficiency of the proposed approach on several regression and classification benchmark datasets including MNIST, CIFAR10, and Airline.
Scalable Gaussian Processes with Billions of Inducing Inputs via Tensor Train Decomposition
by Pavel Izmailov, Alexander Novikov, Dmitry Kropotov
https://arxiv.org/pdf/1710.07324v1.pdf
You must be logged in to reply to this topic.