This repository contains code to determine the learning capability of circuit architectures or ansaetze. It can, thus, be used to reproduce single learning capability values which are depicted in figures of the paper [1].
Install the dependencies via
pip install --upgrade pip
pip install -r requirements.txt
ipython kernel install --user --name=learning_cap
Now you can use you jupyter notebook setup and select the learning_cap kernel.
The notebook learning_capability.ipynb allows to define circuit architectures and to determine their learning capability.
The notebook create_rnd_functions.ipynb enables to create, load and analyse the cross-correlation of truncated Fourier series for a given degree (as described in App. C).
Licensed under the BSD 3-clause license, see LICENSE for details.
In case of any questions, please contact:
Dirk Heimann dirk.heimann@uni-bremen.de or
Gunnar Schönhoff gunnar.schoenhoff@dfki.de
This work was funded by the German Federal Ministry of Education and Research (BMBF) through the project Q$^3$-UP! under project number 40301121 (University of Bremen) and project number 13N15779 (DFKI) administered by the technology center of the Association of German Engineers (VDI) and by the German Federal Ministry of Economics and Climate Protection (BMWK) through the project QuDA-KI under the project numbers 50RA2206A (DFKI) and 50RA2206B (University of Bremen) administered by the German Aerospace Center (DLR).
- [1] Heimann, D., Schönhoff, G., & Kirchner, F. (2022). Learning capability of parametrized quantum circuits. arXiv preprint arXiv:2209.10345.
