GPgym: A Remote Service Platform with Gaussian Process Regression for Online Learning

TL;DR

GPgym proposes a MATLAB-based remote service platform that democratizes Gaussian process regression by removing language barriers for ML deployment. It uses Locally Growing GPs (LoG-GP) with an ARD-SE kernel to enable scalable, local GP inference, configurable via a GUI, and exposed through UDP-based remote services for online learning. The key contributions include a practical integration of LoG-GP with a flexible remote workflow, detailed setup and configuration guidance, and a workflow for running online predictions and simulations, including Monte Carlo testing via a reset command. This facilitates cross-domain adoption of GP techniques in industrial, engineering, and research settings where traditional ML tooling is not readily accessible.

Abstract

Machine learning is now widely applied across various domains, including industry, engineering, and research. While numerous mature machine learning models have been open-sourced on platforms like GitHub, their deployment often requires writing scripts in specific programming languages, such as Python, C++, or MATLAB. This dependency on particular languages creates a barrier for professionals outside the field of machine learning, making it challenging to integrate these algorithms into their workflows. To address this limitation, we propose GPgym, a remote service node based on Gaussian process regression. GPgym enables experts from diverse fields to seamlessly and flexibly incorporate machine learning techniques into their existing specialized software, without needing to write or manage complex script code.

Figures (11)

• Figure 1: GUI.
• Figure 2: UDP for reading.
• Figure 3: UDP for writing.
• Figure 4: Initialize GP models.
• Figure 5: Activate UDP communication.