Dual Parameterization of Sparse Variational Gaussian ProcessesDownload PDF

21 May 2021, 20:50 (modified: 14 Jan 2022, 17:38)NeurIPS 2021 PosterReaders: Everyone
Keywords: Gaussian processes, sparse variational inference, natural gradients
TL;DR: Leveraging dual-parameterization for efficient inference and learning of hyperparameters in sparse variational GP models
Abstract: Sparse variational Gaussian process (SVGP) methods are a common choice for non-conjugate Gaussian process inference because of their computational benefits. In this paper, we improve their computational efficiency by using a dual parameterization where each data example is assigned dual parameters, similarly to site parameters used in expectation propagation. Our dual parameterization speeds-up inference using natural gradient descent, and provides a tighter evidence lower bound for hyperparameter learning. The approach has the same memory cost as the current SVGP methods, but it is faster and more accurate.
Supplementary Material: pdf
Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.
Code: https://github.com/AaltoML/t-SVGP
12 Replies