edahelsinki / gpbound

Error bounds for any regression model using Gaussian processes with gradient information

This repository contains the source code for running the experiments in the paper:

Savvides, Rafael, Hoang Phuc Hau Luu & Kai Puolamäki (2024). Error bounds for any regression model using Gaussian processes with gradient information. AISTATS 2024.

run.sh describes how to run the experiments.

• experiments/ contains all experiment scripts. To run individual experiments, run the corresponding lines in run.sh.
• results/ contains all the results from the experiments.
• data/ contains data sets used in the experiments. Due to file size constraints, yearpredictionmsd and the raw AA, OE62, QM9 are omitted and are instead downloaded from OpenML and Zenodo when running python data/data.py.

Most experiments were run in parallel on a high-performance computing cluster. It is not advised to run all experiments in sequence on a personal computer.