Abstract
This introduction aims to provide readers an intuitive understanding ofGaussian processes regression. Gaussian processes regression (GPR) models havebeen widely used in machine learning applications because their representationflexibility and inherently uncertainty measures over predictions. The paperstarts with explaining mathematical basics that Gaussian processes built onincluding multivariate normal distribution, kernels, non-parametric models,joint and conditional probability. The Gaussian processes regression is thendescribed in an accessible way by balancing showing unnecessary mathematicalderivation steps and missing key conclusive results. An illustrativeimplementation of a standard Gaussian processes regression algorithm isprovided. Beyond the standard Gaussian processes regression, existing softwarepackages to implement state-of-the-art Gaussian processes algorithms arereviewed. Lastly, more advanced Gaussian processes regression models arespecified. The paper is written in an accessible way, thus undergraduatescience and engineering background will find no difficulties in following thecontent.