Precision Aggregated Local Models

TL;DR

This paper introduces Precision Aggregated Local Models (PALM), a novel approach combining local experts and model averaging to achieve fast, accurate, and continuous Gaussian process regression on large datasets.

Contribution

The paper proposes PALM, a new framework that blends divide-and-conquer and model averaging for scalable, continuous Gaussian process modeling, improving speed and accuracy.

Findings

PALM is at least as accurate as LAGP.

PALM can be significantly faster than existing methods.

PALM provides continuous predictive surfaces.

Abstract

Large scale Gaussian process (GP) regression is infeasible for larger data sets due to cubic scaling of flops and quadratic storage involved in working with covariance matrices. Remedies in recent literature focus on divide-and-conquer, e.g., partitioning into sub-problems and inducing functional (and thus computational) independence. Such approximations can be speedy, accurate, and sometimes even more flexible than an ordinary GPs. However, a big downside is loss of continuity at partition boundaries. Modern methods like local approximate GPs (LAGPs) imply effectively infinite partitioning and are thus pathologically good and bad in this regard. Model averaging, an alternative to divide-and-conquer, can maintain absolute continuity but often over-smooths, diminishing accuracy. Here we propose putting LAGP-like methods into a local experts-like framework, blending partition-based speed with model-averaging continuity, as a flagship example of what we call precision aggregated local models (PALM). Using $K$ LAGPs, each selecting $n$ from $N$ total data pairs, we illustrate a scheme that is at most cubic in $n$, quadratic in $K$, and linear in $N$, drastically reducing computational and storage demands. Extensive empirical illustration shows how PALM is at least as accurate as LAGP, can be much faster in terms of speed, and furnishes continuous predictive surfaces. Finally, we propose sequential updating scheme which greedily refines a PALM predictor up to a computational budget.