Comparison of Data Reduction Criteria for Online Gaussian Processes

TL;DR

This paper compares various data reduction criteria for online Gaussian Processes, analyzing their efficiency and effectiveness in managing data size for real-time applications, and offers practical guidelines for their selection.

Contribution

It provides a comprehensive comparison of reduction criteria for online GPs, including new acceptance criteria and practical guidelines for their use.

Findings

Different criteria vary in computational complexity and reduction effectiveness.

Proposed acceptance criteria improve data filtering in online GPs.

Guidelines assist in selecting suitable reduction strategies for real-world tasks.

Abstract

Gaussian Processes (GPs) are widely used for regression and system identification due to their flexibility and ability to quantify uncertainty. However, their computational complexity limits their applicability to small datasets. Moreover in a streaming scenario, more and more datapoints accumulate which is intractable even for Sparse GPs. Online GPs aim to alleviate this problem by e.g. defining a maximum budget of datapoints and removing redundant datapoints. This work provides a unified comparison of several reduction criteria, analyzing both their computational complexity and reduction behavior. The criteria are evaluated on benchmark functions and real-world datasets, including dynamic system identification tasks. Additionally, acceptance criteria are proposed to further filter out redundant datapoints. This work yields practical guidelines for choosing a suitable criterion for an online GP algorithm.