基于量子自编码器的量子机器学习算法研究

机器学习所属计算科学领域，专门分析和解释数据的模式及结构，以实现无 需人工交互即可完成学习、推理和决策等行为的目的，在数据挖掘、计算视觉、自 然语言处理、推荐系统等方面都取得了重要的进展。量子计算利用量子力学独有 的叠加态、纠缠等量子特性，是一种不同于经典计算的革命性计算技术，在人工智 能、药物开发、量子化学、新材料设计以及复杂优化调度等多个方向带来新的革 命。量子机器学习，是机器学习和量子信息的交叉学科，将量子算法整合到机器 学习程序中，主要研究利用量子计算优势提高机器学习对大数据的处理、分析和 挖掘能力，或是在量子计算框架下，提出新型机器学习研究方法，本文对后者的不 同方向进行探索，着重研究量子玻尔兹曼机和量子自编码器这两种类型的量子机 器学习算法，在线路模型量子计算和绝热量子计算框架下，针对不同数据类型和 训练方法所提出的模型进行分析。论文指出基于参数化线路来近似虚时演化过程 的热态近似方法的优势，进而用其实现量子玻尔兹曼机的训练，即参数化量子玻 尔兹曼机;同时将参数化量子玻尔兹曼机应用到具体的问题上，近似条纹图像数 据集的分布，并在开源平台上实现算法的模拟计算，实验结果表明参数化量子玻 尔兹曼机能很好的完成数据分布近似任务，能够推广应用到更复杂数据集、比特 数增大时处理更大维度数据问题上，为实现量子硬件真正应用算法计算提供理论 模型，也为量子深度网络模型提供研究基础。

[1] MOYE W T. ENIAC: the Army-sponsored revolution[J]. US Army Research Laboratory. ftp. arl. army. mil/mike/comphist/96summary/index. html, 1996.
[2] BIAMONTE J, WITTEK P, PANCOTTI N, et al. Quantum machine learning[J/OL]. Nature, 2017, 549(7671): 195–202. https://dx.doi.org/10.1038/nature23474.
[3] REBENTROST P, MOHSENI M, LLOYD S. Quantum Support Vector Machine for Big Data Classification[J/OL]. Physical Review Letters, 2014, 113(13). DOI: 10.1103/physrevlett.113. 130503.
[4] AMIN M H, ANDRIYASH E, ROLFE J, et al. Quantum Boltzmann Machine[J/OL]. Physical Review X, 2018, 8(2). DOI: 10.1103/physrevx.8.021050.
[5] LLOYD S, MOHSENI M, REBENTROST P. Quantum principal component analysis[J/OL]. Nature Physics, 2014, 10(9): 631–633. https://dx.doi.org/10.1038/nphys3029.
[6] PRESKILL J. Quantum Computing in the NISQ era and beyond[J/OL]. Quantum, 2018, 2: 79. DOI: 10.22331/q-2018-08-06-79.
[7] KRAMER M A. Nonlinear principal component analysis using autoassociative neural networks [J/OL]. AIChE Journal, 1991, 37(2): 233-243. https://aiche.onlinelibrary.wiley.com/doi/abs/ 10.1002/aic.690370209. DOI: https://doi.org/10.1002/aic.690370209.
[8] HINTON G E. Boltzmann machine[J/OL]. Scholarpedia, 2007, 2(5): 1668. DOI: 10.4249/sc holarpedia.1668.
[9] ACKLEY D H, HINTON G E, SEJNOWSKI T J. A learning algorithm for boltzmann machines [J/OL]. Cognitive Science, 1985, 9(1): 147-169. https://www.sciencedirect.com/science/articl e/pii/S0364021385800124. DOI: https://doi.org/10.1016/S0364-0213(85)80012-4.
[10] HINTON G E. A Practical Guide to Training Restricted Boltzmann Machines[M/OL]. Berlin, Heidelberg: Springer Berlin Heidelberg, 2012: 599-619. https://doi.org/10.1007/978-3-642-3 5289-8_32.
[11] KHOSHAMAN A, VINCI W, DENIS B, et al. Quantum variational autoencoder[J/OL]. Quan- tum Science and Technology, 2018, 4(1): 014001. http://dx.doi.org/10.1088/2058-9565/aada1f.
[12] VERDON G, BROUGHTON M, BIAMONTE J. A quantum algorithm to train neural networks using low-depth circuits[Z]. 2019.
[13] ZOUFAL C, LUCCHI A, WOERNER S. Variational quantum Boltzmann machines[J]. Quan- tum Machine Intelligence, 2021, 3(1): 1-15.
[14] ZOUFAL C, LUCCHI A, WOERNER S. Quantum generative adversarial networks for learning and loading random distributions[J]. npj Quantum Information, 2019, 5(1): 1-9.
[15] TURING A M, et al. On computable numbers, with an application to the Entscheidungsproblem [J]. J. of Math, 1936, 58(345-363): 5.
[16] KITAEV A Y. Quantum computations: algorithms and error correction[J]. Russian Mathemat- ical Surveys, 1997, 52(6): 1191.
[17] DEUTSCH D, JOZSA R. Rapid solution of problems by quantum computation[J]. Proceed- ings of the Royal Society of London. Series A: Mathematical and Physical Sciences, 1992, 439 (1907): 553-558.
[18] GROVER L K. A fast quantum mechanical algorithm for database search[C]//Proceedings of the twenty-eighth annual ACM symposium on Theory of computing. 1996: 212-219.
[19] BENNETT C H, BERNSTEIN E, BRASSARD G, et al. Strengths and weaknesses of quantum computing[J]. SIAM journal on Computing, 1997, 26(5): 1510-1523.
[20] HARROW A W, HASSIDIM A, LLOYD S. Quantum algorithm for linear systems of equations [J]. Physical review letters, 2009, 103(15): 150502.
[21] ESKANDARPOUR R, GHOSH K, KHODAEI A, et al. Quantum Computing Solution of DC Power Flow[J]. arXiv preprint arXiv:2010.02442, 2020.
[22] ALBASH T, LIDAR D A. Adiabatic quantum computation[J/OL]. Reviews of Modern Physics, 2018, 90(1). http://dx.doi.org/10.1103/RevModPhys.90.015002. DOI: 10.1103/revmodphys.9 0.015002.
[23] MESSIAH A. Quantum mechanics[M]. Courier Corporation, 2014.
[24] FINNILA A, GOMEZ M, SEBENIK C, et al. Quantum annealing: A new method for mini- mizing multidimensional functions[J/OL]. Chemical Physics Letters, 1994, 219(5): 343-348. https://www.sciencedirect.com/science/article/pii/0009261494001170. DOI: https://doi.org/10 .1016/0009-2614(94)00117-0.
[25] KADOWAKI T, NISHIMORI H. Quantum annealing in the transverse Ising model[J/OL]. Phys. Rev. E, 1998, 58: 5355-5363. https://link.aps.org/doi/10.1103/PhysRevE.58.5355.
[26] KIRKPATRICK S, GELATT JR C D, VECCHI M P. Optimization by simulated annealing[J]. science, 1983, 220(4598): 671-680.
[27] AMIN M H, ANDRIYASH E, ROLFE J, et al. Quantum Boltzmann Machine[J/OL]. Physical Review X, 2018, 8(2). http://dx.doi.org/10.1103/PhysRevX.8.021050. DOI: 10.1103/physrevx .8.021050.
[28] ADACHI S H, HENDERSON M P. Application of quantum annealing to training of deep neural networks[J]. arXiv preprint arXiv:1510.06356, 2015.
[29] AÏMEUR E, BRASSARD G, GAMBS S. Machine learning in a quantum world[C]//Conference of the Canadian Society for Computational Studies of Intelligence. Springer, 2006: 431-442.
[30] LEWENSTEIN M. Quantum Perceptrons[J/OL]. Journal of Modern Optics, 1994, 41(12): 2491-2501. https://doi.org/10.1080/09500349414552331.
[31] SHOR P. Algorithms for quantum computation: discrete logarithms and factoring[C/OL]// Proceedings 35th Annual Symposium on Foundations of Computer Science. 1994: 124-134. DOI: 10.1109/SFCS.1994.365700.
[32] NARAYANAN A, MENNEER T. Quantum Artificial Neural Network Architectures and Com- ponents[J/OL]. Inf. Sci., 2000, 128(3–4): 231–255. https://doi.org/10.1016/S0020-0255(00)0 0055-4.
[33] BUHRMAN H, CLEVE R, WATROUS J, et al. Quantum Fingerprinting[J/OL]. Physi- cal Review Letters, 2001, 87(16). http://dx.doi.org/10.1103/PhysRevLett.87.167902. DOI: 10.1103/physrevlett.87.167902.
[34] ANGUITA D, RIDELLA S, RIVIECCIO F, et al. Quantum optimization for training support vector machines[J/OL]. Neural Networks, 2003, 16(5): 763-770. https://www.sciencedirect.co m/science/article/pii/S089360800300087X. DOI: https://doi.org/10.1016/S0893-6080(03)000 87-X.
[35] HARROW A W, HASSIDIM A, LLOYD S. Quantum Algorithm for Linear Systems of Equa- tions[J/OL]. Physical Review Letters, 2009, 103(15). http://dx.doi.org/10.1103/PhysRevLett.1 03.150502. DOI: 10.1103/physrevlett.103.150502.
[36] LLOYD S, MOHSENI M, REBENTROST P. Quantum principal component analysis[J]. Nature Physics, 2014, 10(9): 631-633.
[37] CHATTERJEE R, YU T. Generalized coherent states, reproducing kernels, and quantum sup- port vector machines[J]. arXiv preprint arXiv:1612.03713, 2016.
[38] REBENTROST P, MOHSENI M, LLOYD S. Quantum support vector machine for big data classification[J]. Physical review letters, 2014, 113(13): 130503.
[39] GUARNIERI G, SMIRNE A, VACCHINI B. Quantum regression theorem and non- Markovianity of quantum dynamics[J]. Physical Review A, 2014, 90(2): 022110.
[40] LLOYD S, MOHSENI M, REBENTROST P. Quantum algorithms for supervised and unsuper- vised machine learning[J]. arXiv preprint arXiv:1307.0411, 2013.
[41] DONG D, CHEN C, LI H, et al. Quantum reinforcement learning[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2008, 38(5): 1207-1220.
[42] LOW G H, YODER T J, CHUANG I L. Quantum inference on Bayesian networks[J/OL]. Phys. Rev. A, 2014, 89: 062315. https://link.aps.org/doi/10.1103/PhysRevA.89.062315.
[43] KAPOOR A, WIEBE N, SVORE K. Quantum perceptron models[J]. Advances in neural infor- mation processing systems, 2016, 29.
[44] WIEBE N. Quantum data fitting[C]//APS March Meeting Abstracts: volume 2013. 2013: C27- 006.
[45] WIEBE N, KAPOOR A, SVORE K M. Quantum deep learning[J]. arXiv preprint arXiv:1412.3489, 2014.
[46] DUNJKO V, TAYLOR J M, BRIEGEL H J. Quantum-enhanced machine learning[J]. Physical review letters, 2016, 117(13): 130501.
[47] BIAMONTE J, WITTEK P, PANCOTTI N, et al. Quantum machine learning[J/OL]. Nature, 2017, 549(7671): 195–202. http://dx.doi.org/10.1038/nature23474.
[48] GUPTA S, ZIA R. Quantum neural networks[J]. Journal of Computer and System Sciences, 2001, 63(3): 355-383.
[49] RICKS B, VENTURA D. Training a quantum neural network[J]. Advances in neural informa- tion processing systems, 2003, 16.
[50] SCHULD M, SINAYSKIY I, PETRUCCIONE F. The quest for a Quantum Neural Network [J/OL]. Quantum Inf. Process., 2014, 13(11): 2567-2586. https://doi.org/10.1007/s11128-014 -0809-8.
[51] WAN K H, DAHLSTEN O, KRISTJÁNSSON H, et al. Quantum generalisation of feedforward neural networks[J]. npj Quantum information, 2017, 3(1): 1-8.
[52] CAO Y, GUERRESCHI G G, ASPURU-GUZIK A. Quantum neuron: an elementary building block for machine learning on quantum computers[J]. arXiv preprint arXiv:1711.11240, 2017.
[53] DASKIN A. A simple quantum neural net with a periodic activation function[C]//2018 IEEE International Conference on Systems, Man, and Cybernetics (SMC). IEEE, 2018: 2887-2891.
[54] FARHI E, NEVEN H. Classification with quantum neural networks on near term processors[J]. arXiv preprint arXiv:1802.06002, 2018.
[55] SHAO C. A quantum model for multilayer perceptron[J]. arXiv preprint arXiv:1808.10561, 2018.
[56] BEER K, BONDARENKO D, FARRELLY T, et al. Training deep quantum neural networks[J]. Nature communications, 2020, 11(1): 1-6.
[57] CONG I, CHOI S, LUKIN M D. Quantum convolutional neural networks[J]. Nature Physics, 2019, 15(12): 1273-1278.
[58] BIAMONTE J, WITTEK P, PANCOTTI N, et al. Quantum machine learning[J]. Nature, 2017, 549(7671): 195-202.
[59] ZOUFAL C. Generative Quantum Machine Learning[J]. arXiv preprint arXiv:2111.12738, 2021.
[60] COYLE B, MILLS D, DANOS V, et al. The Born supremacy: quantum advantage and training of an Ising Born machine[J]. npj Quantum Information, 2020, 6(1): 1-11.
[61] FINGERHUTH M, BABEJ T, WITTEK P. Open source software in quantum computing[J]. PloS one, 2018, 13(12): e0208561.
[62] ROMERO J, OLSON J P, ASPURU-GUZIK A. Quantum autoencoders for efficient com- pression of quantum data[J/OL]. Quantum Science and Technology, 2017, 2(4): 045001. http://dx.doi.org/10.1088/2058-9565/aa8072.
[63] LIOU C Y, HUANG J C, YANG W C. Modeling word perception using the Elman network[J]. Neurocomputing, 2008, 71(16-18): 3150-3157.
[64] LIOU C Y, CHENG W C, LIOU J W, et al. Autoencoder for words[J]. Neurocomputing, 2014, 139: 84-96.
[65] GóMEZ-BOMBARELLI R, WEI J N, DUVENAUD D, et al. Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules[J/OL]. ACS Central Science, 2018, 4(2): 268–276. http://dx.doi.org/10.1021/acscentsci.7b00572.
[66] HINTON G E, SEJNOWSKI T J, ACKLEY D H. Boltzmann machines: Constraint satisfac- tion networks that learn[M]. Carnegie-Mellon University, Department of Computer Science Pittsburgh, PA, 1984.
[67] CARREIRA-PERPIÑÁN M A, HINTON G. On Contrastive Divergence Learning[C/OL]// COWELL R G, GHAHRAMANI Z. Proceedings of Machine Learning Research: R5 Pro- ceedings of the Tenth International Workshop on Artificial Intelligence and Statistics. PMLR, 2005: 33-40. https://proceedings.mlr.press/r5/carreira-perpinan05a.html.
[68] HINTON G E, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets[J]. Neural computation, 2006, 18(7): 1527-1554.
[69] KHOSHAMAN A, VINCI W, DENIS B, et al. Quantum variational autoencoder[J/OL]. Quan- tum Science and Technology, 2018, 4(1): 014001. http://dx.doi.org/10.1088/2058-9565/aada1f.
[70] LE CUN Y, FOGELMAN-SOULIÉ F. Modèles connexionnistes de l’apprentissage[J]. Intel- lectica, 1987, 2(1): 114-143.
[71] BOURLARD H, KAMP Y. Auto-association by multilayer perceptrons and singular value de- composition[J]. Biological cybernetics, 1988, 59(4): 291-294.
[72] HINTON G E, ZEMEL R. Autoencoders, Minimum Description Length and Helmholtz Free Energy[C/OL]//COWAN J, TESAURO G, ALSPECTOR J. Advances in Neural Information Processing Systems: volume 6. Morgan-Kaufmann, 1993. https://proceedings.neurips.cc/pap er/1993/file/9e3cfc48eccf81a0d57663e129aef3cb-Paper.pdf.
[73] GÉRON A. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems[M]. ” O’Reilly Media, Inc.”, 2019.
[74] BENGIO Y, LAMBLIN P, POPOVICI D, et al. Greedy layer-wise training of deep networks[J]. Advances in neural information processing systems, 2006, 19.
[75] VINCENT P, LAROCHELLE H, BENGIO Y, et al. Extracting and composing robust features with denoising autoencoders[C]//Proceedings of the 25th international conference on Machine learning. 2008: 1096-1103.
[76] KINGMA D P, WELLING M. Auto-Encoding Variational Bayes[Z]. 2014.
[77] REZENDE D J, MOHAMED S, WIERSTRA D. Stochastic Backpropagation and Approximate Inference in Deep Generative Models[C/OL]//XING E P, JEBARA T. Proceedings of Machine Learning Research: volume 32 Proceedings of the 31st International Conference on Machine Learning. Bejing, China: PMLR, 2014: 1278-1286. https://proceedings.mlr.press/v32/rezend e14.html.
[78] 邱锡鹏. 神经网络与深度学习[M/OL]. 北京: 机械工业出版社, 2020. https://nndl.github.io/.
[79] JANG E, GU S, POOLE B. Categorical reparameterization with gumbel-softmax[J]. arXivpreprint arXiv:1611.01144, 2016.
[80] MAKHZANI A, FREY B J. Pixelgan autoencoders[J]. Advances in Neural Information Pro- cessing Systems, 2017, 30.
[81] HINTON G E. Boltzmann machine[J]. Scholarpedia, 2007, 2(5): 1668.
[82] HINTON G E, SEJNOWSKI T J. Optimal perceptual inference[C]//Proceedings of the IEEE conference on Computer Vision and Pattern Recognition: volume 448. Citeseer, 1983: 448-453.
[83] SALAKHUTDINOV R, HINTON G. Deep Boltzmann Machines[C/OL]//VAN DYK D, WELLING M. Proceedings of Machine Learning Research: volume 5 Proceedings of the Twelth International Conference on Artificial Intelligence and Statistics. Hilton Clearwater Beach Resort, Clearwater Beach, Florida USA: PMLR, 2009: 448-455. https://proceedings. mlr.press/v5/salakhutdinov09a.html.
[84] BENEDETTI M, REALPE-GÓMEZ J, BISWAS R, et al. Estimation of effective temperatures in quantum annealers for sampling applications: A case study with possible applications in deep learning[J/OL]. Phys. Rev. A, 2016, 94: 022308. https://link.aps.org/doi/10.1103/PhysRevA. 94.022308.
[85] KIEFEROVA M, WIEBE N. Tomography and generative data modeling via quantum Boltzmann training[J]. arXiv preprint arXiv:1612.05204, 2016.
[86] WIEBE N, WOSSNIG L. Generative training of quantum Boltzmann machines with hidden units[J]. arXiv preprint arXiv:1905.09902, 2019.
[87] THOMPSON C J. Inequality with applications in statistical mechanics[J]. Journal of Mathe- matical Physics, 1965, 6(11): 1812-1813.
[88] GOLDEN S. Lower Bounds for the Helmholtz Function[J/OL]. Phys. Rev., 1965, 137: B1127- B1128. https://link.aps.org/doi/10.1103/PhysRev.137.B1127.
[89] MARRERO C O, KIEFEROVÁ M, WIEBE N. Entanglement-induced barren plateaus[J]. PRX Quantum, 2021, 2(4): 040316.
[90] MAGNUS W. On the exponential solution of differential equations for a linear operator[J]. Communications on pure and applied mathematics, 1954, 7(4): 649-673.
[91] MCARDLE S, JONES T, ENDO S, et al. Variational ansatz-based quantum simulation of imag- inary time evolution[J]. npj Quantum Information, 2019, 5(1): 1-6.
[92] GUPTA N, ROY A K, DEB B. One-dimensional multiple-well oscillators: A time-dependent quantum mechanical approach[J]. Pramana, 2002, 59(4): 575-583.
[93] AUER J, KROTSCHECK E, CHIN S A. A fourth-order real-space algorithm for solving local Schrödinger equations[J]. The Journal of Chemical Physics, 2001, 115(15): 6841-6846.
[94] RAAB A. On the Dirac–Frenkel/McLachlan variational principle[J/OL]. Chemical Physics Letters, 2000, 319(5): 674-678. https://www.sciencedirect.com/science/article/pii/S0009261 400002001. DOI: https://doi.org/10.1016/S0009-2614(00)00200-1.
[95] KOCZOR B, BENJAMIN S C. Quantum natural gradient generalised to non-unitary circuits [J]. arXiv preprint arXiv:1912.08660, 2019.
[96] MITARAI K, NEGORO M, KITAGAWA M, et al. Quantum circuit learning[J]. Physical Review A, 2018, 98(3): 032309.
[97] DALLAIRE-DEMERS P L, KILLORAN N. Quantum generative adversarial networks[J]. Phys- ical Review A, 2018, 98(1): 012324.
[98] SCHULD M, BERGHOLM V, GOGOLIN C, et al. Evaluating analytic gradients on quantum hardware[J]. Physical Review A, 2019, 99(3): 032331.
[99] MACKAY D J, MAC KAY D J, et al. Information theory, inference and learning algorithms [M]. Cambridge university press, 2003.
[100] MCCLEAN J R, BOIXO S, SMELYANSKIY V N, et al. Barren plateaus in quantum neural network training landscapes[J]. Nature communications, 2018, 9(1): 1-6.
[101] NOH H, YOU T, MUN J, et al. Regularizing deep neural networks by noise: Its interpretation and optimization[J]. Advances in Neural Information Processing Systems, 2017, 30.
[102] GENTILE A A, FLYNN B, KNAUER S, et al. Learning models of quantum systems from experiments[J/OL]. Nature Physics, 2021, 17(7): 837–843. http://dx.doi.org/10.1038/s41567-0 21-01201-7.
[103] BRASPENNING P J, THUIJSMAN F, WEIJTERS A J M M. Artificial neural networks: an introduction to ANN theory and practice: volume 931[M]. Springer Science & Business Media, 1995.
[104] FARHI E, GOLDSTONE J, GUTMANN S. A quantum approximate optimization algorithm [J]. arXiv preprint arXiv:1411.4028, 2014.
[105] HANGLEITERD,ROTHI,NAGAJD,etal.EasingtheMonteCarlosignproblem[J].Science advances, 2020, 6(33): eabb8341.