Data repository for "Minimial-Risk Training Samples for QNN Training from Measurements"

<p>Replication code and experiment result data for training Quantum Neural Networks with entangled data using one-dimensional projectors as observables.
This is the version of the code that was used to generate the experiment results in the related publication. </p>

<p><b>Experiments</b>:<br />
- <code>exp_inf_coeffvariation.py</code>: Trains QNNs using training samples of varying Schmidt rank with fixed vector as Schmidt basis state. Varies the associated Schmidt coefficient.<br />
- <code>exp_inf_random.py</code>: Trains QNNs using random training data.<br/>
</p>

<p>
<b>Experiment results:</b><br />
- <code>exp_inf_coeffvariation.zip</code> and <code>exp_inf_random.zip</code> contain the raw experiment results for both experiments.<br />
- For each combination of controlled variables there is one directory containing the result of all 20 runs of the training process.<br />
- The results for each run are comprised of 3 files: <br />
  - <code>[id]_losses.npy</code>: The loss during the training process<br />
  - <code>[id]_params.npy</code>: The parameters of the QNN after the training process.<br />
  - <code>[id]_V.npy</code>: The trained QNN exported as a 2^4 * 2^4 unitary matrix.
</p>

<p>
<b>Analysis of data</b> (<code>data_extraction.py</code>):<br />
- Computes means and standard deviation of various risk measures and saves the results
</p>

<p>
<b>Plots</b> (<code>plot_obs_risk.py</code>):<br />
- Plots the risk w.r.t. the observable for both experiments based on the analysed data obtained from <code>data_extraction.py</code>.<br />
- Generates <code>plot_coeffvariation.pdf</code> and <code>plot_random.pdf</code>.
</p>