Shot-frugal and Robust quantum kernel classifiers
Abstract: Quantum kernel methods are a candidate for quantum speed-ups in supervised machine learning. The number of quantum measurements N required for a reasonable kernel estimate is a critical resource, both from complexity considerations and because of the constraints of near-term quantum hardware. We emphasize that for classification tasks, the aim is reliable classification and not precise kernel evaluation, and demonstrate that the former is far more resource efficient. Furthermore, it is shown that the accuracy of classification is not a suitable performance metric in the presence of noise and we motivate a new metric that characterizes the reliability of classification. We then obtain a bound for N which ensures, with high probability, that classification errors over a dataset are bounded by the margin errors of an idealized quantum kernel classifier. Using chance constraint programming and the subgaussian bounds of quantum kernel distributions, we derive several Shot-frugal and Robust (ShofaR) programs starting from the primal formulation of the Support Vector Machine. This significantly reduces the number of quantum measurements needed and is robust to noise by construction. Our strategy is applicable to uncertainty in quantum kernels arising from any source of unbiased noise.
- M. Schuld and N. Killoran, Quantum machine learning in feature hilbert spaces, Phys. Rev. Lett. 122, 040504 (2019).
- M. Schuld, Supervised quantum machine learning models are kernel methods,  (2021), arXiv:2101.11020 [quant-ph] .
- J. Kübler, S. Buchholz, and B. Schölkopf, The inductive bias of quantum kernels, Advances in Neural Information Processing Systems 34 (2021).
- Y. Liu, S. Arunachalam, and K. Temme, A rigorous and robust quantum speed-up in supervised machine learning, Nature Physics 17, 1013 (2021).
- C. Cortes and V. Vapnik, Support-vector networks, Machine learning 20, 273 (1995).
- C. J. Burges, A tutorial on support vector machines for pattern recognition, Data mining and knowledge discovery 2, 121 (1998).
- The robust classifiers are constructed by solving a second-order cone program rather than a quadratic program and may thus require more classical resources.
- W. M. Watkins, S. Y.-C. Chen, and S. Yoo, Quantum machine learning with differential privacy, Scientific Reports 13, 2453 (2023).
- L. Banchi, J. Pereira, and S. Pirandola, Generalization in quantum machine learning: A quantum information standpoint, PRX Quantum 2, 040321 (2021).
- V. V. Buldygin and Y. V. Kozachenko, Sub-gaussian random variables, Ukrainian Mathematical Journal 32, 483 (1980).
- V. N. Vapnik, Statistical Learning Theory (Wiley, New York, 1998).
- M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum Information: 10th Anniversary Edition, 10th ed. (Cambridge University Press, USA, 2011).
- P. Massart, Concentration Inequalities and Model Selection, edited by J. Picard (Springer Berlin Heidelberg, 2007).
- I. Steinwart, Sparseness of support vector machines—some asymptotically sharp bounds, Advances in Neural Information Processing Systems 16 (2003).
- A. Charnes, W. W. Cooper, and G. H. Symonds, Cost horizons and certainty equivalents: An approach to stochastic programming of heating oil, Management Science 4, 235 (1958).
- A. Prékopa, On probabilistic constrained programming, in Proceedings of the Princeton Symposium on Mathematical Programming (Princeton University Press, 1970) pp. 113–138.
- A. Prékopa, Stochastic Programming (Kluwer Academic Publishers, Dordrecht, The Netherlands, 1995).
- A. Nemirovski and A. Shapiro, Convex approximations of chance constrained programs, SIAM Journal on Optimization 17, 969 (2007), https://doi.org/10.1137/050622328 .
- A. Jacot, F. Gabriel, and C. Hongler, Neural tangent kernel: Convergence and generalization in neural networks, Advances in neural information processing systems 31 (2018).
- H.-Y. Huang, R. Kueng, and J. Preskill, Predicting many properties of a quantum system from very few measurements, Nature Physics 16, 1050 (2020).
- L. A. Goldberg and H. Guo, The complexity of approximating complex-valued ising and tutte partition functions, computational complexity 26, 765 (2017).
- V. V. Buldigīn and K. K. Moskvichova, Sub-Gaussian norm of a binary random variable, Teor. Ĭmovīr. Mat. Stat. , 28 (2011).
- N. A. Nghiem, S. Y.-C. Chen, and T.-C. Wei, Unified framework for quantum classification, Phys. Rev. Research 3, 033056 (2021).
- G. Boole, The Mathematical Analysis of Logic: Being an Essay Towards a Calculus of Deductive Reasoning, Cambridge Library Collection - Mathematics (Cambridge University Press, 2009).
- S. Diamond and S. Boyd, CVXPY: A Python-embedded modeling language for convex optimization, Journal of Machine Learning Research 17, 1 (2016).
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.
