Quantum State Tomography with Gradient Descent

This repository contains the code required to reproduce the results and plots from the following paper:

📄 IOP Quantum Science and Technology (2025)
📰 arXiv:2503.04526

---

🚀 Overview

This project implements Quantum State Tomography (QST) using different gradient descent methods.
It includes reproducible examples, the full dataset used in the publication, and scripts to regenerate the figures presented in the paper.

---

⚙️ Installation & Usage

Option 1 — Cloning the full repository

This option ensures full reproducibility of the results and provides access to all source code and functions.

1. Clone the repository to obtain the Python modules and notebooks:

   git clone https://github.com/mstorresh/GD-QST.git
   cd GD-QST

2. Create and activate the environment from the provided environment.yml file:

   conda env create -f environment.yml
   conda activate gd_qst_env

⚠️ Important: The code is designed to work with a specific QuTiP version. Please ensure you are using qutip==5.0.0a2 to avoid compatibility issues.

The notebooks in the examples folder are tutorials on how to do the quantum state tomography with the different methods of gradient descent.

The data_and_paper-plots folder contains a Jupyter Notebook (.ipynb) that runs the methods used to generate the plots presented in the paper, along with a ZIP file containing the corresponding data.

Option 2 — Installation via pip

If you only need to use the library (without cloning the full repository), you can install it directly from GitHub.

It is recommended to install it in a new virtual environment:

pip install git+https://github.com/mstorresh/GD-QST.git@pip-gd-qst

Creating a new virtual environment with the environment.yml file is recommended for a clean setup.

The pip-installable version is maintained in the branch pip-gd-qst, currently at version 1.0.0.

📘 Repository Structure

Main Branch

• data_and_paper-plots/ — Contains:

  • The Jupyter Notebook used to reproduce the plots from the paper.
  • A ZIP file with the corresponding data.

• examples/ — Jupyter notebooks demonstrating how to perform QST using various gradient descent methods.

• qst_tec/ — Core Python modules containing the implementation of the gradient descent algorithms and QST routines.

• test/ — Jupyter notebooks to test the gd_qst library installed with pip.

pip-gd-qst Branch

• gd_qst/ — Core Python modules containing the implementation of the gradient descent algorithms and QST routines.

• test/ — Jupyter notebooks to test the gd_qst library installed with pip.

⚠️ Tested environment:

• Python 3.10
• QuTiP 5.0.0a2
• NumPy 1.26.3
• SciPy 1.11.4

💬 Contact

For questions or comments, feel free to contact me directly or use the corresponding author’s email provided in the paper.

📖 Citation

If you use this code or data in your research, please cite:

@article{gaikwad2025gradient,
  title     = {Gradient-descent methods for fast quantum state tomography},
  author    = {Gaikwad, Akshay and Torres, Manuel Sebastian and Ahmed, Shahnawaz and Kockum, Anton Frisk},
  journal   = {Quantum Science and Technology},
  volume    = {10},
  number    = {4},
  pages     = {045055},
  year      = {2025},
  publisher = {IOP Publishing},
  doi       = {10.1088/2058-9565/ae0baa},
  note      = {Published 8 October 2025},
}

Gaikwad, A., Torres, M. S., Ahmed, S., & Kockum, A. F. (2025). Gradient-descent methods for fast quantum state tomography. Quantum Science and Technology, 10(4), 045055.