Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading
Loading data efficiently from classical memories to quantum computers is a key challenge of noisy intermediate-scale quantum computers. Such a problem can be addressed through quantum generative adversarial networks (qGANs), which are noise tolerant and agnostic with respect to data. Tuning a qGAN t...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Quantum Reports |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2624-960X/4/1/6 |
| _version_ | 1797442952724742144 |
|---|---|
| author | Gabriele Agliardi Enrico Prati |
| author_facet | Gabriele Agliardi Enrico Prati |
| author_sort | Gabriele Agliardi |
| collection | DOAJ |
| description | Loading data efficiently from classical memories to quantum computers is a key challenge of noisy intermediate-scale quantum computers. Such a problem can be addressed through quantum generative adversarial networks (qGANs), which are noise tolerant and agnostic with respect to data. Tuning a qGAN to balance accuracy and training time is a hard task that becomes paramount when target distributions are multivariate. Thanks to our tuning of the hyper-parameters and of the optimizer, the training of qGAN reduces, on average, the Kolmogorov–Smirnov statistic of 43–64% with respect to the state of the art. The ability to reach optima is non-trivially affected by the starting point of the search algorithm. A gap arises between the optimal and sub-optimal training accuracy. We also point out that the simultaneous perturbation stochastic approximation (SPSA) optimizer does not achieve the same accuracy as the Adam optimizer in our conditions, thus calling for new advancements to support the scaling capability of qGANs. |
| first_indexed | 2024-03-09T12:49:05Z |
| format | Article |
| id | doaj.art-7ed4b2d37f24489e82804c2f8648014c |
| institution | Directory Open Access Journal |
| issn | 2624-960X |
| language | English |
| last_indexed | 2024-03-09T12:49:05Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Quantum Reports |
| spelling | doaj.art-7ed4b2d37f24489e82804c2f8648014c2023-11-30T22:08:38ZengMDPI AGQuantum Reports2624-960X2022-02-01417510510.3390/quantum4010006Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution LoadingGabriele Agliardi0Enrico Prati1Dipartimento di Fisica, Politecnico di Milano, Piazza Leonardo da Vinci, I-20133 Milano, ItalyIstituto di Fotonica e Nanotecnologie, Consiglio Nazionale delle Ricerche, Piazza Leonardo da Vinci 32, I-20133 Milano, ItalyLoading data efficiently from classical memories to quantum computers is a key challenge of noisy intermediate-scale quantum computers. Such a problem can be addressed through quantum generative adversarial networks (qGANs), which are noise tolerant and agnostic with respect to data. Tuning a qGAN to balance accuracy and training time is a hard task that becomes paramount when target distributions are multivariate. Thanks to our tuning of the hyper-parameters and of the optimizer, the training of qGAN reduces, on average, the Kolmogorov–Smirnov statistic of 43–64% with respect to the state of the art. The ability to reach optima is non-trivially affected by the starting point of the search algorithm. A gap arises between the optimal and sub-optimal training accuracy. We also point out that the simultaneous perturbation stochastic approximation (SPSA) optimizer does not achieve the same accuracy as the Adam optimizer in our conditions, thus calling for new advancements to support the scaling capability of qGANs.https://www.mdpi.com/2624-960X/4/1/6quantum machine learningquantum generative adversarial networksmultivariate quantum distributionsquantum data loadingquantum data encodingquantum finance |
| spellingShingle | Gabriele Agliardi Enrico Prati Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading Quantum Reports quantum machine learning quantum generative adversarial networks multivariate quantum distributions quantum data loading quantum data encoding quantum finance |
| title | Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading |
| title_full | Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading |
| title_fullStr | Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading |
| title_full_unstemmed | Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading |
| title_short | Optimal Tuning of Quantum Generative Adversarial Networks for Multivariate Distribution Loading |
| title_sort | optimal tuning of quantum generative adversarial networks for multivariate distribution loading |
| topic | quantum machine learning quantum generative adversarial networks multivariate quantum distributions quantum data loading quantum data encoding quantum finance |
| url | https://www.mdpi.com/2624-960X/4/1/6 |
| work_keys_str_mv | AT gabrieleagliardi optimaltuningofquantumgenerativeadversarialnetworksformultivariatedistributionloading AT enricoprati optimaltuningofquantumgenerativeadversarialnetworksformultivariatedistributionloading |