Skip to content

Repository files navigation

QumVQD

DOI

Qumode-based Variational Quantum Deflation (QumVQD): a framework for computing electronic and vibrational excited-state energies on bosonic / qumode quantum hardware models. QumVQD maps molecular Hamiltonians onto one or more qumodes and optimizes a SNAP + displacement + beamsplitter ansatz in TensorFlow, recovering ground and excited states through an overlap-penalty based cost function.

Repository layout

helpers/QumVQD_Helpers.py    Core ansatz, VQD loss, optimizer, Hamiltonians
third_party/vibrational_dynamics/   Vibrational Hamiltonian fragmentation code
notebooks/
  Electronic_QumVQD.ipynb    Electronic excited states for H2 / H4
  Vibrational_VQD.ipynb      Vibrational excited states
  Manuscript_Plotting.ipynb  Figures for the manuscript
  Plots/                     Stores plots produced by 
data/                         Data required to run notebooks
pyproject.toml                Package metadata (installable as `QumVQD`)
NOTICE.md                     Third-party attribution
.gitattributes                Handles Git LFS configuration
.gitignore                    Designates which files for Git to ignore
LICENSE.md                    MIT License
THIRD_PARTY_LICENSES.md       Attribution for third-party code   

Installation

Requires Python ≥ 3.9.

Warning: Windows users must use Windows Subsystem for Linux due to dependency on PySCF. Moreover, this repository uses Git LFS for the .hdf5 data files. Install it first (git lfs install), then clone.

git clone https://github.com/sdonglab/QumVQD.git
cd QumVQD
pip install -e .

This installs the helpers and third_party packages along with the dependencies declared in pyproject.toml (QuTiP, TensorFlow, TensorFlow Probability, OpenFermion, openfermionpyscf, scikit-learn, Jupyter).

Quick start

Open one of the notebooks under notebooks/ and run top-to-bottom. Key run parameters at the top of Electronic_QumVQD.ipynb:

Parameter Meaning
molecule "H2" or "H4"
hhdis H–H bond length in Å
num_states Number of states to compute (ground + excited)
ndepth Number of ansatz layers
nfocks List of Fock-space dimensions per qumode, e.g. [70] or [16,16]
beta_factor Weight of the orthogonality penalty in the VQD cost
niter Maximum optimizer iterations per state
threshold Convergence threshold on the loss

Acknowledgments

This repository includes code adapted from:

Both are used under the MIT License. See NOTICE.md, the third_party/ directory, and THIRD_PARTY_LICENSES.md for full attribution and license text.

License

MIT — see LICENSE.

Citations

Paper

Marlon F. Jost, Sijia S. Dong. Excited-State Quantum Chemistry on Qumode-Based Processors via Variational Quantum Deflation. arXiv. 2026. arXiv:2604.13457

@misc{jost2026excitedstatequantumchemistryqumodebased,
      title={Excited-State Quantum Chemistry on Qumode-Based Processors via Variational Quantum Deflation}, 
      author={Marlon F. Jost and Sijia S. Dong},
      year={2026},
      eprint={2604.13457},
      archivePrefix={arXiv},
      primaryClass={quant-ph},
      url={https://arxiv.org/abs/2604.13457}, 
}

Code

Jost, M. and Dong, S. (2026) “QumVQD: Qumode-Based Variational Quantum Deflation Framework”. Zenodo. DOI:10.5281/zenodo.20479296

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages