ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2310.17137
40
7

Large-Scale Gaussian Processes via Alternating Projection

26 October 2023
Kaiwen Wu
Jonathan Wenger
Haydn Thomas Jones
Geoff Pleiss
Jacob R. Gardner
ArXivPDFHTML
Abstract

Training and inference in Gaussian processes (GPs) require solving linear systems with n×nn\times nn×n kernel matrices. To address the prohibitive O(n3)\mathcal{O}(n^3)O(n3) time complexity, recent work has employed fast iterative methods, like conjugate gradients (CG). However, as datasets increase in magnitude, the kernel matrices become increasingly ill-conditioned and still require O(n2)\mathcal{O}(n^2)O(n2) space without partitioning. Thus, while CG increases the size of datasets GPs can be trained on, modern datasets reach scales beyond its applicability. In this work, we propose an iterative method which only accesses subblocks of the kernel matrix, effectively enabling mini-batching. Our algorithm, based on alternating projection, has O(n)\mathcal{O}(n)O(n) per-iteration time and space complexity, solving many of the practical challenges of scaling GPs to very large datasets. Theoretically, we prove the method enjoys linear convergence. Empirically, we demonstrate its fast convergence in practice and robustness to ill-conditioning. On large-scale benchmark datasets with up to four million data points, our approach accelerates GP training and inference by speed-up factors up to 27×27\times27× and 72×72 \times72×, respectively, compared to CG.

View on arXiv
Comments on this paper