Quantum learning: asymptotically optimal classification of qubit states

Pattern recognition is a central topic in Learning Theory with numerous applications such as voice and text recognition, image analysis, computer diagnosis. The statistical set-up in classification is the following: we are given an i.i.d. training set $(X_{1},Y_{1}),... (X_{n},Y_{n})$ where $X_{i}$ represents a feature and $Y_{i}\in \{0,1\}$ is a label attached to that feature. The underlying joint distribution of $(X,Y)$ is unknown, but we can learn about it from the training set and we aim at devising low error classifiers $f:X\to Y$ used to predict the label of new incoming features. Here we solve a quantum analogue of this problem, namely the classification of two arbitrary unknown qubit states. Given a number of `training' copies from each of the states, we would like to `learn' about them by performing a measurement on the training set. The outcome is then used to design mesurements for the classification of future systems with unknown labels. We find the asymptotically optimal classification strategy and show that typically, it performs strictly better than a plug-in strategy based on state estimation. The figure of merit is the excess risk which is the difference between the probability of error and the probability of error of the optimal measurement when the states are known, that is the Helstrom measurement. We show that the excess risk has rate $n^{-1}$ and compute the exact constant of the rate.

[1]  E. Bagan,et al.  Collective versus local measurements in a qubit mixed-state estimation , 2003, quant-ph/0307199.

[2]  Sébastien Gambs,et al.  Quantum classification , 2008, ArXiv.

[3]  J. Kahn,et al.  Local asymptotic normality for qubit states , 2005, quant-ph/0512075.

[4]  J. Kahn,et al.  Optimal Estimation of Qubit States with Continuous Time Measurements , 2006, quant-ph/0608074.

[5]  David A. Cohn,et al.  Active Learning with Statistical Models , 1996, NIPS.

[6]  Giacomo Mauro D'Ariano,et al.  Classical randomness in quantum measurements , 2004, quant-ph/0408115.

[7]  A. Hayashi,et al.  Unambiguous pure-state identification without classical knowledge , 2005, quant-ph/0510015.

[8]  Masahito Hayashi Asymptotic theory of quantum statistical inference : selected papers , 2005 .

[9]  Gerardo Adesso,et al.  Quantum-teleportation benchmarks for independent and identically distributed spin states and displaced thermal states , 2010, 1004.1843.

[10]  Masahide Sasaki,et al.  Quantum learning and universal quantum matching machine , 2002 .

[11]  A. V. D. Vaart,et al.  Asymptotic Statistics: U -Statistics , 1998 .

[12]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[13]  R. Gill,et al.  Optimal full estimation of qubit mixed states , 2005, quant-ph/0510158.

[14]  Richard D. Gill Conciliation of Bayes and Pointwise Quantum State Estimation: Asymptotic information bounds in quantum statistics , 2005 .

[15]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.

[16]  Ruediger Schack,et al.  Unknown Quantum States and Operations, a Bayesian View , 2004, quant-ph/0404156.

[17]  Beck,et al.  Measurement of the Wigner distribution and the density matrix of a light mode using optical homodyne tomography: Application to squeezed states and the vacuum. , 1993, Physical review letters.

[18]  J. M. Radcliffe Some properties of coherent spin states , 1971 .

[19]  A. Jenčová,et al.  Local Asymptotic Normality in Quantum Statistics , 2006, quant-ph/0606213.

[20]  László Györfi,et al.  A Probabilistic Theory of Pattern Recognition , 1996, Stochastic Modelling and Applied Probability.

[21]  G. D’Ariano,et al.  Optimal quantum learning of a unitary transformation , 2009, 0903.0543.

[22]  O. Barndorff-Nielsen,et al.  On quantum statistical inference , 2003, quant-ph/0307189.

[23]  Mark Hillery,et al.  Universal programmable quantum state discriminator that is optimal for unambiguously distinguishing between unknown states. , 2005, Physical review letters.

[24]  J. Kahn,et al.  Local Asymptotic Normality for Finite Dimensional Quantum Systems , 2008, 0804.3876.

[25]  A. V. D. Vaart,et al.  Asymptotic Statistics: Frontmatter , 1998 .

[26]  L. L. Cam,et al.  Asymptotic Methods In Statistical Decision Theory , 1986 .

[27]  H. Yuen Quantum detection and estimation theory , 1978, Proceedings of the IEEE.

[28]  L. Ballentine,et al.  Probabilistic and Statistical Aspects of Quantum Theory , 1982 .

[29]  I. Chuang,et al.  Quantum Computation and Quantum Information: Introduction to the Tenth Anniversary Edition , 2010 .

[30]  Klaus Molmer,et al.  Quantum learning by measurement and feedback , 2008, 0803.1418.

[31]  C. Helstrom Quantum detection and estimation theory , 1969 .

[32]  Separable measurement estimation of density matrices and its fidelity gap with collective protocols. , 2006, Physical review letters.

[33]  A. Hayashi,et al.  Quantum pure-state identification , 2005, quant-ph/0507237.

[34]  Ashutosh Kumar Singh,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2010 .

[35]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[36]  O. Gühne,et al.  03 21 7 2 3 M ar 2 00 6 Scalable multi-particle entanglement of trapped ions , 2006 .

[37]  中澤 真,et al.  Devroye, L., Gyorfi, L. and Lugosi, G. : A Probabilistic Theory of Pattern Recognition, Springer (1996). , 1997 .

[38]  Ueda,et al.  Squeezed spin states. , 1993, Physical review. A, Atomic, molecular, and optical physics.

[39]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[40]  V. Belavkin Generalized uncertainty relations and efficient measurements in quantum systems , 1976, quant-ph/0412030.

[41]  H. Paul,et al.  Measuring the quantum state of light , 1997 .

[42]  A. Tsybakov,et al.  Fast learning rates for plug-in classifiers , 2007, 0708.2321.