002803020 001__ 2803020
002803020 005__ 20240220041335.0
002803020 0248_ $$aoai:cds.cern.ch:2803020$$pcerncds:FULLTEXT$$pcerncds:CERN:FULLTEXT$$pcerncds:CERN
002803020 0247_ $$2DOI$$a10.1088/1742-6596/2438/1/012093
002803020 037__ $$9arXiv$$aarXiv:2203.01007$$cquant-ph
002803020 037__ $$9arXiv:reportnumber$$aMIT-CTP/5400
002803020 035__ $$9arXiv$$aoai:arXiv.org:2203.01007
002803020 035__ $$9Inspire$$aoai:inspirehep.net:2043490$$d2024-02-19T13:39:44Z$$h2024-02-20T03:00:34Z$$mmarcxml$$ttrue$$uhttps://inspirehep.net/api/oai2d
002803020 035__ $$9Inspire$$a2043490
002803020 041__ $$aeng
002803020 100__ $$aBorras, Kerstin$$uDESY, Zeuthen$$uRWTH Aachen U.$$vDeutsches Elektronen-Synchrotron DESY, Platanenallee 6, 15738 Zeuthen, Germany$$vRWTH Aachen University, Templergraben 55, 52062 Aachen, Germany
002803020 245__ $$9arXiv$$aImpact of quantum noise on the training of quantum Generative Adversarial Networks
002803020 269__ $$c2022-03-02
002803020 260__ $$c2023
002803020 300__ $$a6 p
002803020 500__ $$9arXiv$$a6 pages, 5 figures, Proceedings of the 20th International Workshop on
Advanced Computing and Analysis Techniques in Physics Research (ACAT 2021)
002803020 520__ $$9IOP$$aCurrent noisy intermediate-scale quantum devices suffer from various sources of intrinsic quantum noise. Overcoming the effects of noise is a major challenge, for which different error mitigation and error correction techniques have been proposed. In this paper, we conduct a first study of the performance of quantum Generative Adversarial Networks (qGANs) in the presence of different types of quantum noise, focusing on a simplified use case in high-energy physics. In particular, we explore the effects of readout and two-qubit gate errors on the qGAN training process. Simulating a noisy quantum device classically with IBM’s Qiskit framework, we examine the threshold of error rates up to which a reliable training is possible. In addition, we investigate the importance of various hyperparameters for the training process in the presence of different error rates, and we explore the impact of readout error mitigation on the results.
002803020 520__ $$9arXiv$$aCurrent noisy intermediate-scale quantum devices suffer from various sources of intrinsic quantum noise. Overcoming the effects of noise is a major challenge, for which different error mitigation and error correction techniques have been proposed. In this paper, we conduct a first study of the performance of quantum Generative Adversarial Networks (qGANs) in the presence of different types of quantum noise, focusing on a simplified use case in high-energy physics. In particular, we explore the effects of readout and two-qubit gate errors on the qGAN training process. Simulating a noisy quantum device classically with IBM's Qiskit framework, we examine the threshold of error rates up to which a reliable training is possible. In addition, we investigate the importance of various hyperparameters for the training process in the presence of different error rates, and we explore the impact of readout error mitigation on the results.
002803020 540__ $$3preprint$$aCC BY 4.0$$uhttp://creativecommons.org/licenses/by/4.0/
002803020 595_D $$aQ$$d2022-03-09$$sabs
002803020 65017 $$2arXiv$$ahep-ex
002803020 65017 $$2SzGeCERN$$aParticle Physics - Experiment
002803020 65017 $$2arXiv$$aquant-ph
002803020 65017 $$2SzGeCERN$$aGeneral Theoretical Physics
002803020 690C_ $$aCERN
002803020 690C_ $$aARTICLE
002803020 700__ $$aChang, Su [email protected]$$uEcole Polytechnique, Lausanne$$uCERN$$vCERN, 1211 Geneva 23, Switzerland$$vInstitute of Physics, Ecole Polytechnique Fédérale de Lausanne (EPFL), 1015 Lausanne, Switzerland
002803020 700__ $$aFuncke, Lena$$uMIT, Cambridge, CTP$$uIAIFI, Cambridge$$vCenter for Theoretical Physics, Co-Design Center for Quantum Advantage, and NSF AI Institute for Artificial Intelligence and Fundamental Interactions, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
002803020 700__ $$aGrossi, Michele$$uCERN$$vCERN, 1211 Geneva 23, Switzerland
002803020 700__ $$aHartung, Tobias$$uCyprus Inst.$$uBath U.$$vComputation-Based Science and Technology Research Center, The Cyprus Institute, 20 Kavafi Street, 2121 Nicosia, Cyprus$$vDepartment of Mathematical Sciences, 4 West, University of Bath, Claverton Down, Bath BA2 7AY, UK
002803020 700__ $$aJansen, Karl$$uDESY, Zeuthen$$vDeutsches Elektronen-Synchrotron DESY, Platanenallee 6, 15738 Zeuthen, Germany
002803020 700__ $$aKruecker, Dirk$$uDESY, Zeuthen$$vDeutsches Elektronen-Synchrotron DESY, Platanenallee 6, 15738 Zeuthen, Germany
002803020 700__ $$aKühn, Stefan$$uCyprus Inst.$$vComputation-Based Science and Technology Research Center, The Cyprus Institute, 20 Kavafi Street, 2121 Nicosia, Cyprus
002803020 700__ $$aRehm, Florian$$uCERN$$uRWTH Aachen U.$$vCERN, 1211 Geneva 23, Switzerland$$vRWTH Aachen University, Templergraben 55, 52062 Aachen, Germany
002803020 700__ $$aTüysüz, Cenk$$uDESY, Zeuthen$$uHumboldt U., Berlin$$vDeutsches Elektronen-Synchrotron DESY, Platanenallee 6, 15738 Zeuthen, Germany$$vInstitüt für Physik, Humboldt-Universität zu Berlin, Newtonstr. 15, 12489 Berlin, Germany
002803020 700__ $$aVallecorsa, Sofia$$uCERN$$vCERN, 1211 Geneva 23, Switzerland
002803020 773__ $$wC21-11-29
002803020 773__ $$c012093$$n1$$pJ. Phys.: Conf. Ser.$$v2438$$y2023
002803020 8564_ $$82353231$$s1359095$$uhttp://cds.cern.ch/record/2803020/files/cnot_error.png$$y00010 (a) Final relative entropy after qGAN training in the presence of readout noise, as a function of the bit-flip probability $p=p_{01}=p_ {10}$, averaged over $n_{\rm rep}=20$ repetitions, shown with and without error mitigation. (b)~Relative entropy as a function of training epochs, shown with different types of noise, with and without readout-error mitigation.
002803020 8564_ $$82353232$$s774326$$uhttp://cds.cern.ch/record/2803020/files/scan_ro_noise_p01.png$$y00004 : $p=0.1$
002803020 8564_ $$82353233$$s356722$$uhttp://cds.cern.ch/record/2803020/files/qgan_result.png$$y00001 (a) Schematic diagram of the qGAN model, (b) exemplary results of the trained qGAN model applied to a simplified HEP problem using a statevector simulator.
002803020 8564_ $$82353234$$s155452$$uhttp://cds.cern.ch/record/2803020/files/Importance_compare.png$$y00005 Importance of hyperparameters for different strengths of the readout error.
002803020 8564_ $$82353235$$s760948$$uhttp://cds.cern.ch/record/2803020/files/1E-1_compare.png$$y00008 : $p = 0.1$
002803020 8564_ $$82353236$$s495965$$uhttp://cds.cern.ch/record/2803020/files/ent_vs_noise.png$$y00009 (a) Final relative entropy after qGAN training in the presence of readout noise, as a function of the bit-flip probability $p=p_{01}=p_ {10}$, averaged over $n_{\rm rep}=20$ repetitions, shown with and without error mitigation. (b)~Relative entropy as a function of training epochs, shown with different types of noise, with and without readout-error mitigation.
002803020 8564_ $$82353237$$s819611$$uhttp://cds.cern.ch/record/2803020/files/1E-2_compare.png$$y00006 : $p = 0.01$
002803020 8564_ $$82353238$$s772526$$uhttp://cds.cern.ch/record/2803020/files/5E-2_compare.png$$y00007 : $p = 0.05$
002803020 8564_ $$82353239$$s849320$$uhttp://cds.cern.ch/record/2803020/files/scan_ro_noise_p005.png$$y00003 : $p=0.05$
002803020 8564_ $$82353240$$s1547762$$uhttp://cds.cern.ch/record/2803020/files/QGAN_model.png$$y00000 (a) Schematic diagram of the qGAN model, (b) exemplary results of the trained qGAN model applied to a simplified HEP problem using a statevector simulator.
002803020 8564_ $$82353241$$s7569798$$uhttp://cds.cern.ch/record/2803020/files/2203.01007.pdf$$yFulltext
002803020 8564_ $$82353242$$s1038996$$uhttp://cds.cern.ch/record/2803020/files/scan_ro_noise_p001.png$$y00002 : $p=0.01$
002803020 8564_ $$82471751$$s8245765$$uhttp://cds.cern.ch/record/2803020/files/document.pdf$$yFulltext
002803020 960__ $$a13
002803020 962__ $$b2789995$$k012093$$ndaejeon20211129
002803020 980__ $$aConferencePaper
002803020 980__ $$aARTICLE