-
量子人工智能是一个探究人工智能与量子物理交叉的领域: 一方面人工智能的方法和技术可以用来解决量子科学中的问题; 另一方面, 量子计算的发展也可能为人工智能, 尤其是机器学习, 提供新的范式, 极大促进人工智能的发展. 然而, 量子机器学习和经典学习系统对于对抗样本同样具有脆弱性: 在原始数据样本上添加精心制作的微小扰动将很可能导致系统做出错误的预测. 本文介绍经典与量子对抗机器学习的基本概念、原理、以及最新进展. 首先从经典和量子两个方面介绍对抗学习, 通过二维经典伊辛模型和三维手征拓扑绝缘体的对抗样本揭示出经典机器学习在识别物质相时的脆弱性, 同时利用手写字体的对抗样本直观展示出量子分类器的脆弱性. 随后从理论层面上分别阐述经典与量子的“没有免费午餐”定理, 并探讨了量子分类器的普适对抗样本. 最后, 分析并讨论了相应的防御策略. 量子人工智能中对抗学习的研究揭示了量子智能系统潜在的风险以及可能的防御策略, 将对未来量子技术与人工智能的交叉产生深刻影响.Quantum artificial intelligence exploits the interplay between artificial intelligence and quantum physics: on the one hand, a plethora of tools and ideas from artificial intelligence can be adopted to tackle intricate quantum problems; on the other hand, quantum computing could also bring unprecedented opportunities to enhance, speed up, or innovate artificial intelligence. Yet, quantum learning systems, similar to classical ones, may also suffer adversarial attacks: adding a tiny carefully-crafted perturbation to the legitimate input data would cause the systems to make incorrect predictions at a notably high confidence level. In this paper, we introduce the basic concepts and ideas of classical and quantum adversarial learning, as well as some recent advances along this line. First, we introduce the basics of both classical and quantum adversarial learning. Through concrete examples, involving classifications of phases of two-dimensional Ising model and three-dimensional chiral topological insulators, we reveal the vulnerability of classical machine learning phases of matter. In addition, we demonstrate the vulnerability of quantum classifiers with the example of classifying hand-written digit images. We theoretically elucidate the celebrated no free lunch theorem from the classical and quantum perspectives, and discuss the universality properties of adversarial attacks in quantum classifiers. Finally, we discuss the possible defense strategies. The study of adversarial learning in quantum artificial intelligence uncovers notable potential risks for quantum intelligence systems, which would have far-reaching consequences for the future interactions between the two areas.
-
Keywords:
- quantum artificial intelligence /
- quantum adversarial learning /
- quantum classifiers /
- topological phases of mater
[1] Krizhevsky A, Sutskever I, Hinton G E 2012 Proceedings of the 25th International Conference on Neural Information Processing Systems (Volume 1) New York, USA, December 3−8, 2012 p1097
[2] Hinton G, Deng L, Yu D, Dahl G E, Mohamed A, Jaitly N, Senior A, Vanhoucke V, Nguyen P, Sainath T N, Kingsbury B 2012 IEEE Signal Process. Mag. 29 82
Google Scholar
[3] Kononenko I 2001 Artif. Intell. Med. 23 89
Google Scholar
[4] Grigorescu S, Trasnea B, Cocias T, Macesanu G 2020 J. Field Robot 37 362
Google Scholar
[5] LeCun Y, Bengio Y, Hinton G 2015 Nature 521 436
Google Scholar
[6] Jordan M, Mitchell T 2015 Science 349 255
Google Scholar
[7] Silver D, Huang A, Maddison C J, et al. 2016 Nature 529 484
Google Scholar
[8] Silver D, Schrittwieser J, Simonyan K, et al. 2017 Nature 550 354
Google Scholar
[9] Senior A W, Evans R, Jumper J, et al. 2020 Nature 577 706
Google Scholar
[10] Russell S, Norvig P 2020 Artificial Intelligence: A Modern Approach (Hoboken: Pearson) pp1−61
[11] Bishop C 2006 Pattern Recognition and Machine Learning (New York: Springer-Verlag) pp225−284
[12] Nielsen M A, Chuang I L 2010 Quantum Computation and Quantum Information (Cambridge: Cambridge University Press) pp171−352
[13] Arute F, Arya K, Babbush R, et al. 2019 Nature 574 505
Google Scholar
[14] Zhong H S, Wang H, Deng Y H, Chen M C, Peng L C, Luo Y H, Qin J, Wu D, Ding X, Hu Y, Hu P, Yang X Y, Zhang W J, Li H, Li Y, Jiang X, Gan L, Yang G, You L, Wang Z, Li L, Liu N L, Lu C Y, Pan J W 2020 Science 370 1460
[15] Biamonte J, Wittek P, Pancotti N, Rebentrost P, Wiebe N, Lloyd S 2017 Nature 549 195
Google Scholar
[16] Dunjko V, Briegel H J 2018 Rep. Prog. Phys. 81 074001
Google Scholar
[17] Das Sarma S, Deng D L, Duan L M 2019 Phys. Today 72 48
Google Scholar
[18] Carleo G, Troyer M 2017 Science 355 602
Google Scholar
[19] Torlai G, Mazzola G, Carrasquilla J, Troyer M, Melko R, Carleo G 2018 Nat. Phys. 14 447
Google Scholar
[20] Zhang Y H, Zheng P L, Zhang Y, Deng D L 2020 Phys. Rev. Lett. 125 170501
Google Scholar
[21] Deringer V L, Bernstein N, Csányi G, Ben Mahmoud C, Ceriotti M, Wilson M, Drabold D A, Elliott S R 2021 Nature 589 59
Google Scholar
[22] Deng D L 2018 Phys. Rev. Lett. 120 240402
Google Scholar
[23] Deng D L, Li X, Das Sarma S 2017 Phys. Rev. B 96 195145
Google Scholar
[24] Zhang Y, Kim E A 2017 Phys. Rev. Lett. 118 216401
Google Scholar
[25] Carrasquilla J, Melko R G 2017 Nat. Phys. 13 431
Google Scholar
[26] van Nieuwenburg E P L, Liu Y H, Huber S D 2017 Nat. Phys. 13 435
Google Scholar
[27] Wang L 2016 Phys. Rev. B 94 195105
Google Scholar
[28] Broecker P, Carrasquilla J, Melko R G, Trebst S 2017 Sci. Rep. 7 8823
Google Scholar
[29] Ch’ng K, Carrasquilla J, Melko R G, Khatami E 2017 Phys. Rev. X 7 031038
Google Scholar
[30] Wetzel S J 2017 Phys. Rev. E 96 022140
Google Scholar
[31] Hu W, Singh R R P, Scalettar R T 2017 Phys. Rev. E 95 062122
Google Scholar
[32] Zhang Y, Mesaros A, Fujita K, Edkins S D, Hamidian M H, Ch’ng K, Eisaki H, Uchida S, Davis J C S, Khatami E, Kim E-A 2019 Nature 570 484
Google Scholar
[33] Lian W, Wang S T, Lu S, Huang Y, Wang F, Yuan X, Zhang W, Ouyang X, Wang X, Huang X, He L, Chang X, Deng D L, Duan L 2019 Phys. Rev. Lett. 122 210503
Google Scholar
[34] Harrow A W, Hassidim A, Lloyd S 2009 Phys. Rev. Lett. 103 150502
Google Scholar
[35] Lloyd S, Mohseni M, Rebentrost P 2014 Nat. Phys. 10 631
Google Scholar
[36] Dunjko V, Taylor J M, Briegel H J 2016 Phys. Rev. Lett. 117 130501
Google Scholar
[37] Amin M H, Andriyash E, Rolfe J, Kulchytskyy B, Melko R 2018 Phys. Rev. X 8 021050
Google Scholar
[38] Gao X, Zhang Z Y, Duan L M 2018 Sci. Adv. 4 eaat9004
Google Scholar
[39] Lloyd S, Weedbrook C 2018 Phys. Rev. Lett. 121 040502
Google Scholar
[40] Hu L, Wu S H, Cai W, Ma Y, Mu X, Xu Y, Wang H, Song Y, Deng D L, Zou C L, Sun L 2019 Sci. Adv. 5 eaav2761
Google Scholar
[41] Schuld M, Killoran N 2019 Phys. Rev. Lett. 122 040504
Google Scholar
[42] Rebentrost P, Mohseni M, Lloyd S 2014 Phys. Rev. Lett. 113 130503
Google Scholar
[43] Chakraborty A, Alam M, Dey V, Chattopadhyay A, Mukhopadhyay D 2018 arXiv: 1810.00069 [cs, stat]
[44] Biggio B, Roli F 2018 Pattern Recognit. 84 317
Google Scholar
[45] Miller D J, Xiang Z, Kesidis G 2019 arXiv: 1904.06292 [cs, stat]
[46] Szegedy C, Zaremba W, Sutskever I, Bruna J, Erhan D, Goodfellow I, Fergus R 2014 arXiv: 1312.6199 [cs]
[47] Goodfellow I J, Shlens J, Szegedy C 2015 arXiv: 1412.6572 [cs, stat]
[48] Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y 2014 Adv. Neural Inf. Process. Syst. 27 2672
[49] Preskill J 2018 Quantum 2 79
Google Scholar
[50] Lu S, Duan L M, Deng D L 2020 Phys. Rev. Res. 2 033212
Google Scholar
[51] Guan J, Fang W, Ying M 2020 arXiv: 2008.07230 [quant-ph]
[52] Wiebe N, Kumar R S S 2018 New J. Phys. 20 123019
Google Scholar
[53] Liu N, Wittek P 2020 Phys. Rev. A 101 062331
Google Scholar
[54] Gong W, Deng D L 2021 arXiv: 2102.07788 [cond-mat, physics: quant-ph]
[55] Dalvi N, Domingos P, Mausam, Sanghai S, Verma D 2004 Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining Seattle, USA, August 22−25, 2004 p99
[56] [57] Barreno M, Nelson B, Sears R, Joseph A D, Tygar J D 2006 Proceedings of the 2006 ACM Symposium on Information, Computer and Communications Security Taipei, China, March 21−24, 2006 p16
[58] Huang L, Joseph A D, Nelson B, Rubinstein B I P, Tygar J D 2011 Proceedings of the 4th ACM Workshop on Security and Artificial Intelligence Chicago, USA, October 21, 2011 p43
[59] Vorobeychik Y, Kantarcioglu M 2018 Synth. Lect. Artif. Intell. Mach. Learn. 12 1
Google Scholar
[60] Kurakin A, Goodfellow I, Bengio S 2017 arXiv: 1611.01236 [cs, stat]
[61] Ben-Tal A, Ghaoui L E, Nemirovski A 2009 Robust Optimization (Princeton: Princeton University Press) pp1−146
[62] Madry A, Makelov A, Schmidt L, Tsipras D, Vladu A 2018 International Conference on Learning Representations Vancouver, Canada, April 30−May 3, 2018 p1
[63] Kurakin A, Goodfellow I, Bengio S 2017 arXiv: 1607.02533 [cs, stat]
[64] Dong Y, Liao F, Pang T, Su H, Zhu J, Hu X, Li J 2018 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Salt Lake City, USA, June 18−23, 2018 p9185
[65] Hsu Y T, Li X, Deng D L, Das Sarma S 2018 Phys. Rev. Lett. 121 245701
Google Scholar
[66] Rodriguez-Nieva J F, Scheurer M S 2019 Nat. Phys. 15 790
Google Scholar
[67] Zhang P, Shen H, Zhai H 2018 Phys. Rev. Lett. 120 066401
Google Scholar
[68] Huembeli P, Dauphin A, Wittek P 2018 Phys. Rev. B 97 134109
Google Scholar
[69] Rem B S, Käming N, Tarnowski M, Asteria L, Fläschner N, Becker C, Sengstock K, Weitenberg C 2019 Nat. Phys. 15 917
Google Scholar
[70] Bohrdt A, Chiu C S, Ji G, Xu M, Greif D, Greiner M, Demler E, Grusdt F, Knap M 2019 Nat. Phys. 15 921
Google Scholar
[71] Jiang S, Lu S, Deng D L 2019 arXiv: 1910.13453 [cond-mat, physics: quant-ph]
[72] Neupert T, Santos L, Ryu S, Chamon C, Mudry C 2012 Phys. Rev. B 86 035125
Google Scholar
[73] Deng D L, Wang S T, Duan L M 2014 Phys. Rev. A 90 041601
Google Scholar
[74] Finlayson S G, Bowers J D, Ito J, Zittrain J L, Beam A L, Kohane I S 2019 Science 363 1287
Google Scholar
[75] Ren K, Zheng T, Qin Z, Liu X 2020 Engineering 6 346
Google Scholar
[76] Goodfellow I, Bengio Y, Courville A 2016 Deep Learning (Cambridge: The MIT Press) pp98−165
[77] Sutskever I, Vinyals O, Le Q V 2014 Proceedings of the 27th International Conference on Neural Information Processing Systems (Volume 2) Montréal, Canada, December 8−13, 2014 p3104
[78] Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez A N, Kaiser L, Polosukhin I 2017 arXiv: 1706.03762 [cs]
[79] Wolpert D H, Macready W G 1997 IEEE Trans. Evol. Comput. 1 67
Google Scholar
[80] Shalev-Shwartz S, Ben-David S 2014 Understanding Machine Learning: From Theory to Algorithms (New York: Cambridge University Press) pp36−41
[81] Poland K, Beer K, Osborne T J 2020 arXiv: 2003.14103 [quant-ph]
[82] Schroff F, Kalenichenko D, Philbin J 2015 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Boston, USA, June 7−12, 2015 p815
[83] Walters M 2015 arXiv: 1508.05448 [math-ph]
[84] Schwabl F 2006 Statistical Mechanics (Berlin Heidelberg: Springer-Verlag) pp1−20
[85] Dohmatob E 2019 arXiv: 1810.04065 [cs, stat]
[86] Sharif M, Bhagavatula S, Bauer L, Reiter M K 2016 Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security Vienna, Austria, October 24−28, 2016 p1528
[87] Papernot N, McDaniel P, Goodfellow I, Jha S, Celik Z B, Swami A 2017 Proceedings of the 2017 ACM on Asia Conference on Computer and Communications Security Abu Dhabi, UAE, April 2−6 2017 p506
[88] Ledoux M 2001 The Concentration of Measure Phenomenon (Providence: American Mathematical Society) pp1–21
[89] Hoeffding W 1963 J. Am. Stat. Assoc. 58 13
Google Scholar
[90] Sharma K, Cerezo M, Holmes Z, Cincio L, Sornborger A, Coles P J 2020 arXiv: 2007.04900 [quant-ph]
[91] Du Y, Hsieh M H, Liu T, Tao D, Liu N 2020 arXiv: 2003.09416 [quant-ph]
[92] Tramèr F, Kurakin A, Papernot N, Goodfellow I, Boneh D, McDaniel P 2017 arXiv: 1705.07204 [cs, stat]
[93] Papernot N, McDaniel P, Wu X, Jha S, Swami A 2016 37th IEEE Symposium on Security and Privacy San Jose, USA, May 23−25, 2016 p582
[94] Samangouei P, Kabkab M, Chellappa R 2018 arXiv: 1805.06605 [cs, stat]
[95] Dallaire-Demers P L, Killoran N 2018 Phys. Rev. A 98 012324
Google Scholar
[96] Huang K, Wang Z A, Song C, Xu K, Li H, Wang Z, Guo Q, Song Z, Liu Z-B, Zheng D, Deng D L, Wang H, Tian J G, Fan H 2020 arXiv: 2009.12827 [quant-ph]
[97] Zeng J, Wu Y, Liu J G, Wang L, Hu J 2019 Phys. Rev. A 99 052306
Google Scholar
[98] Anderson P W 1967 Phys. Rev. Lett. 18 1049
Google Scholar
[99] Deng D L, Pixley J H, Li X, Das Sarma S 2015 Phys. Rev. B 92 220201
Google Scholar
-
图 1 量子与经典对抗学习示意图 输入的原始熊猫图像样本可以编码为经典或量子数据, 分类器(包含变分量子线路或人工神经网络)能够以非常高的准确率识别出熊猫; 但添加少量精心制作的噪声后, 同一分类器将以非常高的置信度把轻微修改过的熊猫图像错误分类为长臂猿
Fig. 1. A schematic illustration of quantum and classical adversarial learning. The image of a panda can be encoded as classical or quantum data. A classifier, which uses either variational quantum circuits or classical artificial neural networks, can successfully identify the image as a panda with the state-of-the-art accuracy. However, adding a small amount of carefully crafted noise will cause the same classifier to misclassify the slightly modified image into a gibbon with a notably high confidence.
图 2 机器学习物质相中的对抗样本 (a)一个原始的经典二维伊辛模型铁磁相的自旋构型; (b)被分类器错误识别成顺磁相的对抗样本, 其相对于(a)只改变了一个自旋; (c)一个原始的三维手征拓扑绝缘体的拓扑相样本; (d)被分类器错误识别成其他相的对抗样本, 其相对于(c)只有肉眼难以识别的细微差别
Fig. 2. Adversarial examples in machine learning phases of matter: (a) A legitimate sample of the spin configuration in the ferromagnetic phase of the two-dimensional (2D) classical Ising model; (b) an adversarial example misclassified as the paramagnetic phase, which only differs from the original legitimate one shown in (a) by a single pixel; (c) a legitimate sample of the topological phase of three-dimensional (3D) chiral topological insulators; (d) an adversarial example misclassified as the other phase, which only differs from the original legitimate one shown in (c) by a tiny amount of noises that are imperceptible to human eyes.
图 3 量子分类器在识别MNIST中手写字体图片时的对抗样本 (a)经过无差别攻击, 量子分类器以极高置信度将数字7, 9分别识别成9, 7, 即使对抗样本和初始样本的差别非常微小; (b)通过针对性攻击, 量子分类器将把对抗样本预测为给定错误标签, 尽管对抗样本和初始样本相差无几
Fig. 3. Adversarial examples in quantum learning of MNIST hand-written images: (a) After untargeted attacks, the quantum classifier will misclassify the images of digit 7 (9) as digit 9 (7) with notably high confidence, although the differences between the adversarial and legitimate images are tiny; (b) after targeted attack, the quantum classifier will misclassify the adversarial examples into the category with the targeted label, even though the adversarial and legitimate images only differ slightly from each other.
-
[1] Krizhevsky A, Sutskever I, Hinton G E 2012 Proceedings of the 25th International Conference on Neural Information Processing Systems (Volume 1) New York, USA, December 3−8, 2012 p1097
[2] Hinton G, Deng L, Yu D, Dahl G E, Mohamed A, Jaitly N, Senior A, Vanhoucke V, Nguyen P, Sainath T N, Kingsbury B 2012 IEEE Signal Process. Mag. 29 82
Google Scholar
[3] Kononenko I 2001 Artif. Intell. Med. 23 89
Google Scholar
[4] Grigorescu S, Trasnea B, Cocias T, Macesanu G 2020 J. Field Robot 37 362
Google Scholar
[5] LeCun Y, Bengio Y, Hinton G 2015 Nature 521 436
Google Scholar
[6] Jordan M, Mitchell T 2015 Science 349 255
Google Scholar
[7] Silver D, Huang A, Maddison C J, et al. 2016 Nature 529 484
Google Scholar
[8] Silver D, Schrittwieser J, Simonyan K, et al. 2017 Nature 550 354
Google Scholar
[9] Senior A W, Evans R, Jumper J, et al. 2020 Nature 577 706
Google Scholar
[10] Russell S, Norvig P 2020 Artificial Intelligence: A Modern Approach (Hoboken: Pearson) pp1−61
[11] Bishop C 2006 Pattern Recognition and Machine Learning (New York: Springer-Verlag) pp225−284
[12] Nielsen M A, Chuang I L 2010 Quantum Computation and Quantum Information (Cambridge: Cambridge University Press) pp171−352
[13] Arute F, Arya K, Babbush R, et al. 2019 Nature 574 505
Google Scholar
[14] Zhong H S, Wang H, Deng Y H, Chen M C, Peng L C, Luo Y H, Qin J, Wu D, Ding X, Hu Y, Hu P, Yang X Y, Zhang W J, Li H, Li Y, Jiang X, Gan L, Yang G, You L, Wang Z, Li L, Liu N L, Lu C Y, Pan J W 2020 Science 370 1460
[15] Biamonte J, Wittek P, Pancotti N, Rebentrost P, Wiebe N, Lloyd S 2017 Nature 549 195
Google Scholar
[16] Dunjko V, Briegel H J 2018 Rep. Prog. Phys. 81 074001
Google Scholar
[17] Das Sarma S, Deng D L, Duan L M 2019 Phys. Today 72 48
Google Scholar
[18] Carleo G, Troyer M 2017 Science 355 602
Google Scholar
[19] Torlai G, Mazzola G, Carrasquilla J, Troyer M, Melko R, Carleo G 2018 Nat. Phys. 14 447
Google Scholar
[20] Zhang Y H, Zheng P L, Zhang Y, Deng D L 2020 Phys. Rev. Lett. 125 170501
Google Scholar
[21] Deringer V L, Bernstein N, Csányi G, Ben Mahmoud C, Ceriotti M, Wilson M, Drabold D A, Elliott S R 2021 Nature 589 59
Google Scholar
[22] Deng D L 2018 Phys. Rev. Lett. 120 240402
Google Scholar
[23] Deng D L, Li X, Das Sarma S 2017 Phys. Rev. B 96 195145
Google Scholar
[24] Zhang Y, Kim E A 2017 Phys. Rev. Lett. 118 216401
Google Scholar
[25] Carrasquilla J, Melko R G 2017 Nat. Phys. 13 431
Google Scholar
[26] van Nieuwenburg E P L, Liu Y H, Huber S D 2017 Nat. Phys. 13 435
Google Scholar
[27] Wang L 2016 Phys. Rev. B 94 195105
Google Scholar
[28] Broecker P, Carrasquilla J, Melko R G, Trebst S 2017 Sci. Rep. 7 8823
Google Scholar
[29] Ch’ng K, Carrasquilla J, Melko R G, Khatami E 2017 Phys. Rev. X 7 031038
Google Scholar
[30] Wetzel S J 2017 Phys. Rev. E 96 022140
Google Scholar
[31] Hu W, Singh R R P, Scalettar R T 2017 Phys. Rev. E 95 062122
Google Scholar
[32] Zhang Y, Mesaros A, Fujita K, Edkins S D, Hamidian M H, Ch’ng K, Eisaki H, Uchida S, Davis J C S, Khatami E, Kim E-A 2019 Nature 570 484
Google Scholar
[33] Lian W, Wang S T, Lu S, Huang Y, Wang F, Yuan X, Zhang W, Ouyang X, Wang X, Huang X, He L, Chang X, Deng D L, Duan L 2019 Phys. Rev. Lett. 122 210503
Google Scholar
[34] Harrow A W, Hassidim A, Lloyd S 2009 Phys. Rev. Lett. 103 150502
Google Scholar
[35] Lloyd S, Mohseni M, Rebentrost P 2014 Nat. Phys. 10 631
Google Scholar
[36] Dunjko V, Taylor J M, Briegel H J 2016 Phys. Rev. Lett. 117 130501
Google Scholar
[37] Amin M H, Andriyash E, Rolfe J, Kulchytskyy B, Melko R 2018 Phys. Rev. X 8 021050
Google Scholar
[38] Gao X, Zhang Z Y, Duan L M 2018 Sci. Adv. 4 eaat9004
Google Scholar
[39] Lloyd S, Weedbrook C 2018 Phys. Rev. Lett. 121 040502
Google Scholar
[40] Hu L, Wu S H, Cai W, Ma Y, Mu X, Xu Y, Wang H, Song Y, Deng D L, Zou C L, Sun L 2019 Sci. Adv. 5 eaav2761
Google Scholar
[41] Schuld M, Killoran N 2019 Phys. Rev. Lett. 122 040504
Google Scholar
[42] Rebentrost P, Mohseni M, Lloyd S 2014 Phys. Rev. Lett. 113 130503
Google Scholar
[43] Chakraborty A, Alam M, Dey V, Chattopadhyay A, Mukhopadhyay D 2018 arXiv: 1810.00069 [cs, stat]
[44] Biggio B, Roli F 2018 Pattern Recognit. 84 317
Google Scholar
[45] Miller D J, Xiang Z, Kesidis G 2019 arXiv: 1904.06292 [cs, stat]
[46] Szegedy C, Zaremba W, Sutskever I, Bruna J, Erhan D, Goodfellow I, Fergus R 2014 arXiv: 1312.6199 [cs]
[47] Goodfellow I J, Shlens J, Szegedy C 2015 arXiv: 1412.6572 [cs, stat]
[48] Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y 2014 Adv. Neural Inf. Process. Syst. 27 2672
[49] Preskill J 2018 Quantum 2 79
Google Scholar
[50] Lu S, Duan L M, Deng D L 2020 Phys. Rev. Res. 2 033212
Google Scholar
[51] Guan J, Fang W, Ying M 2020 arXiv: 2008.07230 [quant-ph]
[52] Wiebe N, Kumar R S S 2018 New J. Phys. 20 123019
Google Scholar
[53] Liu N, Wittek P 2020 Phys. Rev. A 101 062331
Google Scholar
[54] Gong W, Deng D L 2021 arXiv: 2102.07788 [cond-mat, physics: quant-ph]
[55] Dalvi N, Domingos P, Mausam, Sanghai S, Verma D 2004 Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining Seattle, USA, August 22−25, 2004 p99
[56] [57] Barreno M, Nelson B, Sears R, Joseph A D, Tygar J D 2006 Proceedings of the 2006 ACM Symposium on Information, Computer and Communications Security Taipei, China, March 21−24, 2006 p16
[58] Huang L, Joseph A D, Nelson B, Rubinstein B I P, Tygar J D 2011 Proceedings of the 4th ACM Workshop on Security and Artificial Intelligence Chicago, USA, October 21, 2011 p43
[59] Vorobeychik Y, Kantarcioglu M 2018 Synth. Lect. Artif. Intell. Mach. Learn. 12 1
Google Scholar
[60] Kurakin A, Goodfellow I, Bengio S 2017 arXiv: 1611.01236 [cs, stat]
[61] Ben-Tal A, Ghaoui L E, Nemirovski A 2009 Robust Optimization (Princeton: Princeton University Press) pp1−146
[62] Madry A, Makelov A, Schmidt L, Tsipras D, Vladu A 2018 International Conference on Learning Representations Vancouver, Canada, April 30−May 3, 2018 p1
[63] Kurakin A, Goodfellow I, Bengio S 2017 arXiv: 1607.02533 [cs, stat]
[64] Dong Y, Liao F, Pang T, Su H, Zhu J, Hu X, Li J 2018 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Salt Lake City, USA, June 18−23, 2018 p9185
[65] Hsu Y T, Li X, Deng D L, Das Sarma S 2018 Phys. Rev. Lett. 121 245701
Google Scholar
[66] Rodriguez-Nieva J F, Scheurer M S 2019 Nat. Phys. 15 790
Google Scholar
[67] Zhang P, Shen H, Zhai H 2018 Phys. Rev. Lett. 120 066401
Google Scholar
[68] Huembeli P, Dauphin A, Wittek P 2018 Phys. Rev. B 97 134109
Google Scholar
[69] Rem B S, Käming N, Tarnowski M, Asteria L, Fläschner N, Becker C, Sengstock K, Weitenberg C 2019 Nat. Phys. 15 917
Google Scholar
[70] Bohrdt A, Chiu C S, Ji G, Xu M, Greif D, Greiner M, Demler E, Grusdt F, Knap M 2019 Nat. Phys. 15 921
Google Scholar
[71] Jiang S, Lu S, Deng D L 2019 arXiv: 1910.13453 [cond-mat, physics: quant-ph]
[72] Neupert T, Santos L, Ryu S, Chamon C, Mudry C 2012 Phys. Rev. B 86 035125
Google Scholar
[73] Deng D L, Wang S T, Duan L M 2014 Phys. Rev. A 90 041601
Google Scholar
[74] Finlayson S G, Bowers J D, Ito J, Zittrain J L, Beam A L, Kohane I S 2019 Science 363 1287
Google Scholar
[75] Ren K, Zheng T, Qin Z, Liu X 2020 Engineering 6 346
Google Scholar
[76] Goodfellow I, Bengio Y, Courville A 2016 Deep Learning (Cambridge: The MIT Press) pp98−165
[77] Sutskever I, Vinyals O, Le Q V 2014 Proceedings of the 27th International Conference on Neural Information Processing Systems (Volume 2) Montréal, Canada, December 8−13, 2014 p3104
[78] Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez A N, Kaiser L, Polosukhin I 2017 arXiv: 1706.03762 [cs]
[79] Wolpert D H, Macready W G 1997 IEEE Trans. Evol. Comput. 1 67
Google Scholar
[80] Shalev-Shwartz S, Ben-David S 2014 Understanding Machine Learning: From Theory to Algorithms (New York: Cambridge University Press) pp36−41
[81] Poland K, Beer K, Osborne T J 2020 arXiv: 2003.14103 [quant-ph]
[82] Schroff F, Kalenichenko D, Philbin J 2015 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Boston, USA, June 7−12, 2015 p815
[83] Walters M 2015 arXiv: 1508.05448 [math-ph]
[84] Schwabl F 2006 Statistical Mechanics (Berlin Heidelberg: Springer-Verlag) pp1−20
[85] Dohmatob E 2019 arXiv: 1810.04065 [cs, stat]
[86] Sharif M, Bhagavatula S, Bauer L, Reiter M K 2016 Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security Vienna, Austria, October 24−28, 2016 p1528
[87] Papernot N, McDaniel P, Goodfellow I, Jha S, Celik Z B, Swami A 2017 Proceedings of the 2017 ACM on Asia Conference on Computer and Communications Security Abu Dhabi, UAE, April 2−6 2017 p506
[88] Ledoux M 2001 The Concentration of Measure Phenomenon (Providence: American Mathematical Society) pp1–21
[89] Hoeffding W 1963 J. Am. Stat. Assoc. 58 13
Google Scholar
[90] Sharma K, Cerezo M, Holmes Z, Cincio L, Sornborger A, Coles P J 2020 arXiv: 2007.04900 [quant-ph]
[91] Du Y, Hsieh M H, Liu T, Tao D, Liu N 2020 arXiv: 2003.09416 [quant-ph]
[92] Tramèr F, Kurakin A, Papernot N, Goodfellow I, Boneh D, McDaniel P 2017 arXiv: 1705.07204 [cs, stat]
[93] Papernot N, McDaniel P, Wu X, Jha S, Swami A 2016 37th IEEE Symposium on Security and Privacy San Jose, USA, May 23−25, 2016 p582
[94] Samangouei P, Kabkab M, Chellappa R 2018 arXiv: 1805.06605 [cs, stat]
[95] Dallaire-Demers P L, Killoran N 2018 Phys. Rev. A 98 012324
Google Scholar
[96] Huang K, Wang Z A, Song C, Xu K, Li H, Wang Z, Guo Q, Song Z, Liu Z-B, Zheng D, Deng D L, Wang H, Tian J G, Fan H 2020 arXiv: 2009.12827 [quant-ph]
[97] Zeng J, Wu Y, Liu J G, Wang L, Hu J 2019 Phys. Rev. A 99 052306
Google Scholar
[98] Anderson P W 1967 Phys. Rev. Lett. 18 1049
Google Scholar
[99] Deng D L, Pixley J H, Li X, Das Sarma S 2015 Phys. Rev. B 92 220201
Google Scholar
计量
- 文章访问数: 8718
- PDF下载量: 567
- 被引次数: 0