Data repository for "Minimial-Risk Training Samples for QNN Training from Measurements" (doi:10.18419/darus-4113)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

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

doi:10.18419/darus-4113

DaRUS

2024-10-08

1

Mandl, Alexander; Bechtold, Marvin; Barzen, Johanna; Leymann, Frank, 2024, "Data repository for "Minimial-Risk Training Samples for QNN Training from Measurements"", https://doi.org/10.18419/darus-4113, DaRUS, V1

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

doi:10.18419/darus-4113

Mandl, Alexander (Universität Stuttgart)

Bechtold, Marvin (Universität Stuttgart)

Barzen, Johanna (Universität Stuttgart)

Leymann, Frank (Universität Stuttgart)

01MQ22007B

01MQ22009B

DaRUS

Mandl, Alexander

Mandl, Alexander

2024-03-25

https://doi.org/10.18419/darus-4113

Computer and Information Science, Quantum Neural Network, Quantum Computing, Quantum Entanglement

Artificial Intelligence and Machine Learning Methods, Hardware Systems and Architectures for Information Technology and Artificial Intelligence, Quantum Engineering Systems

<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 /> &nbsp;&nbsp;- <code>[id]_losses.npy</code>: The loss during the training process<br /> &nbsp;&nbsp;- <code>[id]_params.npy</code>: The parameters of the QNN after the training process.<br /> &nbsp;&nbsp;- <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>

2024-01-15-2024-03-01

Simulation data

Mandl, Alexander; Barzen, Johanna; Bechtold, Marvin; Leymann, Frank, 2024, "Minimial-Risk Training Samples for QNN Training from Measurements"

Mandl, Alexander; Barzen, Johanna; Bechtold, Marvin; Leymann, Frank, 2024, "Minimial-Risk Training Samples for QNN Training from Measurements"

Label:

data_extended.py

Text:

Functions for generating training data of various structures.

text/x-python

Label:

data_extraction.py

Text:

Analyzes raw experiment data to compute means and standard deviation of various metrics.

text/x-python

Label:

experiment.py

Text:

General experiment code. Trains networks and outputs the trained network.

text/x-python

Label:

exp_inf_coeffvariation.py

Text:

Experiment entry point for the experiment in Section 4.2/Figure 1.

text/x-python

Label:

exp_inf_random.py

Text:

Experiment entry point for the experiment in Section 4.3/Figure 2.

text/x-python

Label:

exp_setup.py

Text:

General setup code for experiments - is imported in the experiment entry points.

text/x-python

Label:

plot_coeffvariation.pdf

Text:

Figure 1: Average risk of QNNs that are trained with training samples of varying Schmidt rank r and varying coefficient c_1 for the Schmidt basis state |γ⟩ = U^†|o⟩.

application/pdf

Label:

plot_obs_risks.py

Text:

Main entry points for plotting figures from the paper.

text/x-python

Label:

plot_random.pdf

Text:

Figure 2: Average risk of QNNs that are trained with training samples of varying Schmidt rank r comprised of randomly sampled Schmidt basis vectors.

application/pdf

Label:

pt_train.py

Text:

The QNN training routines that are used in the experiments.

text/x-python

Label:

requirements.txt

Text:

Python requirements for experiments.

text/plain

Label:

risk.py

Text:

Utility functions for computing risk.

text/x-python

Label:

sampling_error_extension.py

Text:

Functions/Classes required for training with observables/training by sampling from measurements.

text/x-python

Label:

utils.py

Text:

General utility functions.

text/x-python

Label:

exp_inf_coeffvariation.zip

Text:

Raw experiment results for the experiments shown in Figure 1.

application/zip

Label:

exp_inf_random.zip

Text:

Raw experiment results for the experiments shown in Figure 2.

application/zip

Label:

cuda_qnn.py

Text:

Collection of various QNN implementations. For the paper the implementation UnitaryParametrization is used.

text/x-python

Label:

qnn.py

Text:

Structure of QNN implementations.

text/x-python

Label:

quantum_gates.py

Text:

Collection of various quantum gates implemented using pytorch for the QNN implementations.

text/x-python

Label:

unitaries.zip

Text:

Unitary matrices (as pytorch files) that were used as target operators in the experiments.

application/zip