DAO-GP Drift Aware Online Non-Linear Regression Gaussian-Process

February 09, 2026

Reading time: 2 minute

...

📝 Original Info

Title: DAO-GP Drift Aware Online Non-Linear Regression Gaussian-Process
ArXiv ID: 2512.08879
Date: 2025-12-09
Authors: Mohammad Abu-Shaira, Ajita Rattani, Weishi Shi

📝 Abstract

Real-world datasets often exhibit temporal dynamics characterized by evolving data distributions. Disregarding this phenomenon, commonly referred to as concept drift, can significantly diminish a model's predictive accuracy. Furthermore, the presence of hyperparameters in online models exacerbates this issue. These parameters are typically fixed and cannot be dynamically adjusted by the user in response to the evolving data distribution. Gaussian Process (GP) models offer powerful non-parametric regression capabilities with uncertainty quantification, making them ideal for modeling complex data relationships in an online setting. However, conventional online GP methods face several critical limitations, including a lack of drift-awareness, ...

📄 Full Content

In today's highly dynamic, data-centric environment, marked by a continuous influx of information from diverse sources such as IoT devices, financial markets, and user interactions, the demand for real-time learning and adaptive capabilities has become increasingly critical [1]. However, traditional batch learning falls short due to assumptions of stationary distributions, need for complete datasets, and lack of continuous updates. These inherent drawbacks severely impede their efficacy in dynamic environments. For instance, models trained on historical stock market data often fail to respond to real-time shifts in market sen

…(Content truncated for length.)

📄 Read Full PDF on ArXiv