This repository contains the implementation of the the Graph Gaussian Process model introduced in 'Bayesian semi-supervised learning with graph Gaussian processes'.
@inproceedings{ng2018gaussian,
title={Bayesian semi-supervised learning with graph Gaussian processes},
author={Ng, Yin Cheng and Colombo, Nicolo and Silva, Ricardo},
booktitle={Advances in Neural Information Processing Systems},
year={2018}
}
It is an updated version of the semi-supervised classification model of the original implementation accompanying the paper: yincheng/GGP. It was updated to use GPflow 2 and TensorFlow 2.2.