Approximations to Gaussian processes (GPs) based on inducing variables, combined with variational inference techniques, enable stateof-the-art sparse approaches to infer GPs at scale through mini-batch-based learning. In this work, we address one limitation of sparse GPs, which is due to the challenge in dealing with a large number of inducing variables without imposing a special structure on the inducing inputs. In particular, we introduce a novel hierarchical prior, which imposes sparsity on the set of inducing variables. We treat our model variationally, and we experimentally show considerable computational gains compared to standard sparse GPs when sparsity on the inducing variables is realized considering the nearest inducing inputs of a random mini-batch of the data. We perform an extensive experimental validation that demonstrates the effectiveness of our approach compared to the state-of-the-art. Our approach enables the possibility to use sparse GPs using a large number of inducing points without incurring a prohibitive computational cost.
Sparse within sparse gaussian processes using neighbor information
Submitted to ArXiV, 12 November 2020
© EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in Submitted to ArXiV, 12 November 2020 and is available at :
PERMALINK : https://www.eurecom.fr/publication/6401