Lightweight Distributed Gaussian Process Regression for Online Machine Learning

• 成果类型: Article
• 署名作者: Yuan, Zhenyuan; Zhu, Minghui
• 署名单位: Pennsylvania Commonwealth System of Higher Education (PCSHE); Pennsylvania State University; Pennsylvania State University - University Park
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2024.3351555
• 发表日期: 2024
• 页码: 3928-3943
• 关键词: Gaussian processes; Prediction algorithms; kernel; Training data; training; Servers; Approximation algorithms; Distributed algorithms; Machine Learning

摘要

In this article, we study the problem where a group of agents aims to collaboratively learn a common static latent function through streaming data. We propose alight weight distributed Gaussian process regression (GPR)algorithm that is cognizant of agents' limited capabilities in communication, computation, and memory. Each agent independently runs agent-based GPR using local streaming data to predict test points of interest; then, the agents collaboratively execute distributed GPR to obtain global predictions over a common sparse set of test points; finally, each agent fuses results from distributed GPR with agent-based GPR to refine its predictions. By quantifying the transient and steady-state performances in predictive variance and error, we show that limited inter agent communication improves learning performances in the sense of Pareto. Monte Carlo simulation is conducted to evaluate the developed algorithm.