Code for "Safe Bayesian Optimization for Uncertain Correlations Matrices in Linear Models of Co-Regionalization"

This repository contains the source code and supplementary material for the paper:

Jannis O. Lübsen and Annika Eichler, "Safe Bayesian Optimization for Uncertain Correlations Matrices in Linear Models of Co-Regionalization" Submitted to the IFAC World Congress 2026.

The code allows for the reproduction of all tables and figures presented in the manuscript.

Safe Bayesian Optimization for Uncertain Correlations Matrices in Linear Models of Co-Regionalization

This repository contains the code required to reproduce all tables and figures presented in the manuscript: "Safe Bayesian Optimization for Uncertain Correlations Matrices in Linear Models of Coregionalization."

Prerequisites

The codebase was developed and tested in the following environment:

• OS: Ubuntu 24.04.2 LTS
• Language: Python 3.12.8

Installation

To set up the environment, follow these steps:

1. Clone the repository:

   git clone https://github.com/TUHH-ICS/2025-code-Safe-Bayesian-Optimization-for-Uncertain-Correlations-Matrices-in-LMCs.git

2. Navigate to the project directory:

   cd ./2025-code-safe-bayesian-optimization-for-uncertain-correlations-matrices-in-linear-models-of-coregionalization

3. Install dependencies:

   pip install -r requirements.txt

Reproducing Results

1. Running the Optimization Algorithms

Multitask Bayesian Optimization: To run the multitask algorithm, execute the rkhs_opt.py script. You can specify the model type (ICM or LMC) using the model_type parameter inside the rkhs_opt.py script (default is LMC).

# Run with default LMC model
python rkhs_opt.py

Single Task Bayesian Optimization: For the single-task case, execute the rkhs_opt.py script.

python rkhs_opt_ST.py

2. Generating Plots

To generate the plots used in the manuscript:

1. Navigate to the plot_scripts directory
2. Open and run the Jupyter Notebook generate_plots.ipynb