SKIing on Simplices: Kernel Interpolation on the Permutohedral Lattice for Scalable Gaussian Processes
arXiv:2106.06695
Abstract
State-of-the-art methods for scalable Gaussian processes use iterative algorithms, requiring fast matrix vector multiplies (MVMs) with the covariance kernel. The Structured Kernel Interpolation (SKI) framework accelerates these MVMs by performing efficient MVMs on a grid and interpolating back to the original space. In this work, we develop a connection between SKI and the permutohedral lattice used for high-dimensional fast bilateral filtering. Using a sparse simplicial grid instead of a dense rectangular one, we can perform GP inference exponentially faster in the dimension than SKI. Our approach, Simplex-GP, enables scaling SKI to high dimensions, while maintaining strong predictive performance. We additionally provide a CUDA implementation of Simplex-GP, which enables significant GPU acceleration of MVM based inference.
International Conference on Machine Learning (ICML), 2021
References in corpus (6)
- Language Models are Few-Shot Learners
- Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
- Scalable Variational Gaussian Process Classification
- Kernel Interpolation for Scalable Structured Gaussian Processes (KISS-GP)
- MCMC for Variationally Sparse Gaussian Processes
- Permutohedral Lattice CNNs