GBSGE

This repository contains python code for approximating the Gaussian expectation problems using GBS-I, GBS-P, and MC methods.

Features

• Generate Gaussian expectation problems.
• Simulate and compare GBS-I, GBS-P, and MC methods.
• Create figures and tables in the companion paper.

Installation

git clone https://github.com/sshanshans/GBEGE.git
pip install -e ./

Usage

1. Generate a Gaussian Expectation Problem

To generate a Gaussian expectation problem, use one of the following scripts:

• script/model_haf.py
• script/model_hafsq.py

---

2. Simulate Comparisons

To simulate and compare the performance of GBS-I, GBS-P, and MC methods, run:

• script/run_est_haf.py
• script/run_est_hafsq.py

---

3. Generate Figures

To produce figures for analyzing the convergence behaviors, use the script:

• script/make_fig_haf.py
• script/make_fig_hafsq.py

---

4. Generate Tables

To reproduce tables displayed in the paper, run one of the example scripts:

• script/example1.py
• script/example2.py
• script/example3.py
• script/example4.py

License

This software is licensed under the GPL-3.0 License. See the LICENSE file for details.

Citations

If you use this software in your research, please cite the associated paper:

Using Gaussian Boson Samplers for Approximate Gaussian Expectation Problems
Jørgen Ellegaard Andersen, Shan Shan (2025)

Contributions

Contributions are welcome! Please submit a pull request or open an issue if you encounter bugs or have suggestions.