betaFeedback

WorksPhysics and Astronomy

Triple descent in quantum linear regression

Abstract

Quantum linear regression models, such as quantum extreme learning machines and quantum fidelity kernels, are trained on features that can only be estimated from finitely many measurement shots. Sampling noise is usually treated as a nuisance that more shots suppress; we show instead that it can fundamentally reshape the instabilities of the learning problem. As the number of measurement outcomes grows, the prediction error displays a triple-descent profile, with two peaks of fundamentally different origins. The first, classical, peak occurs when the measurement first becomes informationally complete, is a direct analogue of double descent in classical linear regression, and is caused by ill-conditioning of the measurement itself. This peak affects predictions only through their variance, and is thus suppressed by increasing the number of shots at inference. The second, quantum, peak occurs beyond informational completeness when the number of outcomes matches the number of training states: shot noise populates unphysical outcome combinations and the regression fits these pure-noise features with large weights, corrupting the learned model itself. More training or test shots cannot remove this quantum peak, which disappears only in the ideal limit of exact training features. Using random-matrix theory, we derive the shape of both peaks analytically, accounting also for ridge regularization. Our results show that treating quantum features as exactly known overestimates classical instabilities while missing genuinely quantum ones.

Introduction

Most quantum machine learning (QML) models share a common structure: an input quantum state is measured after evolution through some quantum dynamics, and a classical model is then trained on features extracted from the measurement outcomes [1, 2, 3, 4]. These features are never known exactly, since expectation values, outcome probabilities, and state overlaps can only be estimated from finitely many shots. While the resulting sampling noise is usually treated as a nuisance that more shots can make arbitrarily small, this picture is incomplete: sampling noise can produce effects that do not vanish with more shots, and reducing it can even worsen performance in some cases. In a companion paper [5], for example, we showed that standard training methods can end up fitting the sampling noise itself, a form of quantum overfitting that worsens as the training statistics increase.

Quantum linear regression — We study these effects in quantum linear regression (QLR) models [5], a class of QML protocols that includes quantum extreme learning machines [6, 7, 8, 9, 10, 11], quantum reservoir computing [12, 13, 14], and quantum fidelity kernels [3, 4, 15, 16]. In QLR, a quantum device produces the features as expectation values of fixed measurement operators, and the prediction is a linear combination of these features whose coefficients are fitted classically by standard regression [17, 18, 19, 20].

Spectral origins of double descent — Fitting these coefficients generally involves inverting the matrix of features measured on the training states, an operation which can amplify errors along its nearly singular directions. In ridgeless and random-feature regression, this mechanism underlies double descent phenomena [21, 22, 23]: as features are added, the test error peaks when their number matches the number of training examples, and then decreases again. The peak arises because the feature matrix is square at that point, and random matrix theory shows that square random matrices typically have singular values close to zero [24, 25, 26]. In shadow tomography, the same mechanism makes observable estimation from poorly conditioned measurements particularly sensitive to shot noise [27]. Spectral analyses have recently extended this picture to QML [28, 29, 30, 31], revealing double-descent phenomena that do not originate from shot noise and persist even when the features are known exactly.

Triple descent from finite sampling — Here, we show that sampling noise adds a second peak to this picture, producing a triple descent as the number of measurement outcomes is varied. Unlike previously identified double-descent peaks, this new peak has an intrinsically stochastic origin: more training statistics do not suppress it — in the test variance, its height actually grows and then saturates to a value depending on the dimension — and it disappears only in the theoretical limit of exactly known training features. Unlike the peak at informational completeness, which affects only the test variance when the target is reconstructed exactly, the new peak can also distort the learned observable itself, and is therefore not removed by increasing the statistics at inference. We focus our analysis on the simplest type of quantum features: the outcome probabilities of a fixed measurement performed after a fixed quantum evolution. This choice loses little generality, since the mechanisms we identify rely only on two properties shared by all QLR models: the exact features depend linearly on the input state, and they are estimated from finite samples. Learning a target observable then means reconstructing its expectation value from the outcome probabilities, using ntrn_{\mathrm{tr}} training states labeled by their target values.

Two spectral instabilities — More precisely, as the number of measurement outcomes noutn_{\mathrm{out}} grows, the prediction error peaks twice, as illustrated schematically in Figure 1, and the two peaks have different origins. We call a peak classical if it is caused by ill-conditioning of the exact feature matrix, typically inherited from the measurement itself, and quantum if it is caused by nearly singular directions that sampling fluctuations create. In both cases, the peak appears because a random matrix that is close to square is likely to be nearly singular, as described by the Marchenko–Pastur law. We also map out the full landscape of possible peaks arising at the various completeness and interpolation thresholds in Figure 2.

Figure 1. Triple descent in quantum linear regression. Pictorial representation of the MSE versus the number of measurement outcomes noutn_{\mathrm{out}}, for ntr>d2n_{\mathrm{tr}} > d^2 training states and various test statistics NN. Top: With exact training probabilities (N=∞N = \infty), the error exhibits a classical peak at the threshold for informational completeness, nout=d2n_{\mathrm{out}} = d^2, caused by ill-conditioning of the measurement frame. Middle: With large but finite training statistics (N≫1N \gg1 shots per training state), sampling fluctuations populate additional feature directions absent from the exact probabilities. These generate a second, quantum peak at the interpolation threshold, nout=ntrn_{\mathrm{out}} = n_{\mathrm{tr}}, producing the triple-descent profile. Bottom: At low training statistics, shot noise suppresses the classical peak through effective regularization, and the quantum peak becomes dominant. The labels indicate regions where the measurement is informationally complete (IC), not informationally complete (NO-IC), underparameterized (UP), and overparameterized (OP).

Figure 2. Overview of classical and quantum peaks. Schematic representation of the regression regimes and the instabilities that separate them, for generic random POVMs and training states. With high probability, the measurement is informationally complete for nout≥d2n_{\mathrm{out}} \ge d^2, while the training states span operator space for ntr≥d2n_{\mathrm{tr}} \ge d^2. For ntr>d2n_{\mathrm{tr}} > d^2, increasing noutn_{\mathrm{out}} we cross we cross two peak: a classical one at nout=d2n_{\mathrm{out}} = d^2, caused by ill-conditioning of the measurement and affecting the test variance even with exact training features (§ 3), and a quantum peak at nout=ntrn_{\mathrm{out}} = n_{\mathrm{tr}}, caused by finite training shot-noise (section IV). Increasing noutn_{\mathrm{out}} at ntr>d2n_{\mathrm{tr}} > d^2 thus causes a triple descent phenomenon, as also illustrated in Figure 1. Similarly, for nout>d2n_{\mathrm{out}} > d^2, increasing ntrn_{\mathrm{tr}} crosses a classical interpolation threshold at ntr=d2n_{\mathrm{tr}} = d^2, where label noise can produce a peak even with exact features, followed by the quantum peak at ntr=noutn_{\mathrm{tr}} = n_{\mathrm{out}}. When both nout,ntr<d2n_{\mathrm{out}}, n_{\mathrm{tr}} < d^2 the interpolation peak nout=ntrn_{\mathrm{out}} = n_{\mathrm{tr}} changes nature and becomes classical, and remains even with exact training and test features and no label noise, being caused by inaccessible components of the target observable (§ 3.2). This is analogous to the peak observed in [32]. The classical peak at ntr>d2n_{\mathrm{tr}} > d^2 is caused by a divergence in the test variance and is thus visible only accounting for finite test statistics; the classical peak at ntr=d2n_{\mathrm{tr}} = d^2 is caused by label noise and is thus visible only for γ>0\gamma> 0; the quantum peak at nout=ntrn_{\mathrm{out}} = n_{\mathrm{tr}} is instead caused by stochastic instabilities in the estimated regression weights, and is thus visible only accounting for finite training statistics N<∞N < \infty.

Peak at informational completeness — The first peak occurs at informational completeness, nout=d2n_{\mathrm{out}} = d^2 with dd the Hilbert-space dimension, the point at which exact outcome probabilities first suffice to reconstruct the input state. This peak is classical and is the direct counterpart of the double descent observed in classical linear regression [21] and quantum kernel methods [29]. Formally, it originates from the near-singularity of the measurement frame operator [27], a superoperator built from the POVM that describes which operator directions the measurement resolves, and how well. For the random POVMs we consider, the POVM elements form an effectively random nout×d2n_{\mathrm{out}} \times d^2 matrix, so the frame-operator spectrum follows the Marchenko–Pastur law, and becomes poorly conditioned exactly where the POVM first becomes informationally complete. Because this ill-conditioning belongs to the measurement, the peak persists with exact training probabilities, and would appear even if the measurement were known exactly and the target were reconstructed with the optimal unbiased linear estimator [27]; suitably chosen symmetric measurements, however, can avoid it: for instance symmetric informationally complete POVMs, whose measurement frame operator is proportional to the identity on traceless operators [33]. At low training statistics, shot noise acts as a regularizer and dampens the peak [5, 32], while at high training statistics its height saturates to a level set by the exact features. Because this peak stems from large exact regression coefficients, its effect on predictions is confined to their variance, and more shots at inference suppress it.

Classical peak from label noise — For nout>d2n_{\mathrm{out}} > d^2, a further classical peak can occur at ntr=d2n_{\mathrm{tr}} = d^2 when the training labels are noisy, even with exact measurement probabilities. Near this threshold, the training states become nearly linearly dependent: certain combinations ∑iciρi\sum_i c_i \rho_i almost vanish, while the corresponding combinations of noisy labels generally do not. Least-squares regression attempts to fit these fluctuations by introducing large components in the reconstructed observable along the poorly constrained operator directions. The resulting amplification of label noise produces a peak in the prediction error that persists even for N=M=∞N = M = \infty. In the absence of label noise, the near-dependencies are exactly reflected in the labels, and this instability does not occur.

Peak at interpolation — At the interpolation threshold nout=ntrn_{\mathrm{out}} = n_{\mathrm{tr}}, the classification depends on whether the measurement is informationally complete. For ntr<d2n_{\mathrm{tr}} < d^2, the exact training-feature matrix can itself become poorly conditioned and the resulting peak is classical, as also observed in [32]. For ntr>d2n_{\mathrm{tr}} > d^2, the peak is quantum: exact probabilities span at most d2d^2 dimensions, but training-shot fluctuations populate the remaining directions. The noise-only part of the empirical feature matrix becomes square at nout=ntrn_{\mathrm{out}} = n_{\mathrm{tr}}; its small singular values make the least-squares fit amplify test-shot noise and perturb the reconstructed observable. Consequently, adding outcomes beyond informational completeness can worsen predictions even as the measurement frame becomes better conditioned.

Finite-size behavior and limits — At finite Hilbert-space dimension, the peaks are rounded and their heights fluctuate across measurements and training records. The divergences arise in distinct high-dimensional limits, studied below. In particular, the typical quantum test-variance peak saturates as the training-shot budget NN grows at fixed dimension, although the peak is absent when the training probabilities are supplied exactly. The order of these limiting operations is therefore essential.

Analytical approach and implications — We make all these observations quantitative by analyzing the frame operator with random-matrix methods and by separating, in the empirical feature matrix, the directions carried by the training states from those generated only by sampling. The resulting analysis proves that both peaks exist and yields their shape analytically, both with and without ridge regularization, in good agreement with extensive numerical simulations. More broadly, these results show that the spectrum of an ideal quantum feature map is only part of the learning problem: treating the features as exactly known misses the quantum peak entirely and overestimates the classical one. A reliable analysis must therefore treat training and test statistics separately, and account both for the constraints that quantum states and measurements impose on the features and for the finite records from which they are estimated. In practice, measurement design, training-set size, and shot budget cannot be chosen independently: whether adding outcomes helps or hurts depends on how noutn_{\rm out} compares with both d2d^2 and ntrn_{\rm tr}.

We first define the regression protocol and separate prediction error into its ideal-test and test-variance contributions in § 2. § 3 identifies the classical peak at informational completeness. Section IV analyzes the quantum peak at interpolation above informational completeness, including its training-shot and ridge dependence. The classical interpolation peak below informational completeness is treated separately in § 3.2. We close with implications and open questions in section VI. For readability, theorem statements in the main text are informal; precise formulations and proofs are in the Supplemental Material.

Quantum linear regression and prediction error

In this section we set up the learning problem and isolate the contributions of training and test measurement statistics to the prediction error. This separation will let us identify the two thresholds characterizing the triple-descent phenomenon and the nature of the quantum and classical peaks. A schematic of the measurement protocol we consider is given in Figure 3.

Figure 3. Learning protocol. Each training state ρi\rho_i is measured NN times with the POVM μ\mu, and the resulting outcome frequencies are used to train a linear least-squares regressor with weights ww. The regressor then predicts the target value for a new test state σ\sigma from the outcome frequencies of MM measurements of σ\sigma. For any ww, the prediction y^\hat{y} averaged over test shot noise is Tr⁡(O~σ)\operatorname{Tr}(\widetilde{O}\sigma), with the effective observable O~=∑bwbμb\widetilde{O} = \sum_b w_b\mu_b. When the training labels are yi=Tr⁡(Oρi)y_i = \operatorname{Tr}(O\rho_i) and O∈span⁡{μb}O \in\operatorname{span}\{\mu_b\}, which always holds for informationally complete POVMs, and the training states span the operator space, the learned weights give O~≈O\widetilde{O} \approx O for large NN.

Quantum features and empirical regression

We consider dd-dimensional states ρ\rho measured with a fixed positive-operator-valued measure (POVM) μ=(μa)a=1nout\mu= (\mu_a)_{a=1}^{n_{\rm out}} with noutn_{\rm out} outcomes, where μa≥0\mu_a \ge0 and ∑aμa=Id\sum_a \mu_a = I_d. Here μ\mu is the effective POVM [6], which accounts

for both the physical measurement and any quantum evolution applied before it. The outcome probabilities p(ρ)p(\rho), with components pa(ρ)=Tr⁡(μaρ)p_a(\rho) = \operatorname{Tr}(\mu_a\rho), serve as features for a classical linear model trained to predict the expectation value y(ρ)=Tr⁡(Oρ)y(\rho) = \operatorname{Tr}(O\rho) of a Hermitian target observable OO. We assume that the training labels are known exactly, so that finite statistics affect learning only through the features.

Rank constraint and informational completeness — For a training set of ntrn_{\rm tr} states ρ=(ρi)i=1ntr\rho= (\rho_i)_{i=1}^{n_{\rm tr}}, we collect the labels yi=y(ρi)y_i = y(\rho_i) in the vector yy and the exact probabilities in the matrix P∈Rnout×ntrP \in\mathbb{R}^{n_{\rm out}\times n_{\rm tr}}, whose ii-th column is p(ρi)p(\rho_i). The probability matrix factorizes as

P=MμTMρ,rank⁡P≤d2,(1)P = M_\mu^{T}M_\rho,\qquad\operatorname{rank} P \le d^2, \tag*{(1)}

where the columns of Mμ∈Rd2×noutM_\mu\in\mathbb{R}^{d^2\times n_{\rm out}} and Mρ∈Rd2×ntrM_\rho\in\mathbb{R}^{d^2\times n_{\rm tr}} are the effects μa\mu_a and the states ρi\rho_i, vectorized in an orthonormal basis of the d2d^2-dimensional real vector space of Hermitian operators on Cd\mathbb{C}^d; this rank is bounded because all effects and states live in this d2d^2-dimensional space. Unless stated otherwise, we assume that the training states span the space of Hermitian operators, which requires ntr≥d2n_{\mathrm{tr}} \ge d^2. The bound is then saturated, rank⁡P=d2\operatorname{rank} P=d^2, precisely when μ\mu is informationally complete, i.e., when its effects also span the space. For an informationally complete POVM, every target observable lies in span⁡({μb})\operatorname{span}(\{\mu_b\}) and can therefore be reconstructed from the measurement data. For other POVMs, the components of O\mathcal{O} outside span⁡({μb})\operatorname{span}(\{\mu_b\}) are inaccessible: whenever such components are present, the prediction error remains nonzero even with unlimited training resources.

Finite-shot feature matrix — In an experiment, however, PP is not directly available: its ii-th column is estimated as n^i/N\hat{n}_i/N, where the count vector n^i\hat{n}_i records the outcomes of NN measurements of ρi\rho_i and is multinomially distributed,

n^i∣ρi∼Mult⁡(N,pi),Σi=diag⁡pi−pipiT,(2)\hat{n}_i \mid\rho_i \sim\operatorname{Mult}(N,p_i), \qquad\Sigma_i=\operatorname{diag}p_i-p_i p_i^{\mathsf{T}}, \tag*{(2)}

References

  1. [1]J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Quantum machine learning, Nature 549, 195 (2017).DOI
  2. [2]M. Schuld and N. Killoran, Quantum machine learning in feature Hilbert spaces, Physical review letters 122, 040504 (2019).DOI
  3. [3]V. Havlíček, A. D. Córcoles, K. Temme, A. W. Harrow, A. Kandala, J. M. Chow, and J. M. Gambetta, Supervised learning with quantum-enhanced feature spaces, Nature 567, 209 (2019).DOI
  4. [4]H.-Y. Huang, M. Broughton, M. Mohseni, R. Babbush, S. Boixo, H. Neven, and J. R. McClean, Power of data in quantum machine learning, Nature communications 12, 2631 (2021).
  5. [5]G. L. Monaco, S. Lorenzo, A. Ferraro, M. Paternostro, G. M. Palma, and L. Innocenti, Who can sample forever? Shot noise effects in quantum linear regression (2026), arXiv:2610.12015 [quant-ph].
  6. [6]L. Innocenti, S. Lorenzo, I. Palmisano, A. Ferraro, M. Paternostro, and G. M. Palma, Potential and limitations of quantum extreme learning machines, Communications Physics 6, 118 (2023).DOI
  7. [7]W. Xiong, G. Facelli, M. Sahebi, O. Agnel, T. Chotibut, S. Thanasip, and Z. Holmes, On fundamental aspects of quantum extreme learning machines, Quantum Machine Intelligence 7, 20 (2025).DOI
  8. [8]M. Vetrano, G. Lo Monaco, L. Innocenti, S. Lorenzo, and G. M. Palma, State estimation with quantum extreme learning machines beyond the scrambling time, npj Quantum Information 11, 20 (2025).DOI
  9. [9]G. Lo Monaco, M. Bertini, S. Lorenzo, and G. M. Palma, Quantum extreme learning of molecular potential energy surfaces and force fields, Machine Learning: Science and Technology 5, 035014 (2024).DOI
  10. [10]A. Suprano, D. Zia, L. Innocenti, S. Lorenzo, V. Cimini, T. Giordani, I. Palmisano, E. Polino, N. Spagnolo, F. Scia-rino, et al., Experimental property reconstruction in a photonic quantum extreme learning machine, Physical Review Letters 132, 160802 (2024).DOI
  11. [11]D. Zia, L. Innocenti, G. Minati, S. Lorenzo, A. Suprano, R. Di Bartolo, N. Spagnolo, T. Giordani, V. Cimini, G. M. Palma, et al., Quantum reservoir computing for photonic entanglement witnessing, Science Advances 11, eady7987 (2025).DOI
  12. [12]K. Fujii and K. Nakajima, Harnessing disordered quantum dynamics for machine learning, Physical Review Applied 8, 024030 (2017), arXiv:1602.08159 [quant-ph].DOI
  13. [13]S. Ghosh, A. Opala, M. Matuszewski, T. Paterek, and T. C. H. Liew, Quantum reservoir processing, npj Quantum Information 5, 35 (2019).DOI
  14. [14]P. Mujal, R. Martínez-Peña, J. Nokkala, J. García-Beni, G. L. Giorgi, M. C. Soriano, and R. Zambrini, Opportunities in Quantum Reservoir Computing and Extreme Learning Machines, Advanced Quantum Technologies 4, 2100027 (2021).DOI
  15. [15]M. Schuld, Supervised quantum machine learning models are kernel methods (2021), arXiv:2101.11020 [quant-ph, stat].arxiv.org/abs/2101.11020
  16. [16]S. Jerbi, L. J. Fiderer, H. Poulsen Nautrup, J. M. Kübler, H. J. Briegel, and V. Dunjko, Quantum machine learning beyond kernel methods, Nature Communications 14, 517DOI
  17. [17]B. Schölkopf and A. J. Smola, Learning with kernels: support vector machines, regularization, optimization, and beyond (the MIT Press, 2001).
  18. [18]T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, Springer Series in Statistics (Springer, 2009).
  19. [19]G.-B. Huang, Q.-Y. Zhu, and C.-K. Siew, Extreme learning machine: theory and applications, Neurocomputing 70, 489 (2006).DOI
  20. [20]H. Jaeger and H. Haas, Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication, science 304, 78 (2004).
  21. [21]T. Hastie, A. Montanari, S. Rosset, and R. J. Tibshirani, Surprises in high-dimensional ridgeless least squares interpolation, The Annals of Statistics 50, 949 (2022).DOI
  22. [22]M. Belkin, D. Hsu, S. Ma, and S. Mandal, Reconciling modern machine-learning practice and the classical bias–variance trade-off, Proceedings of the National Academy of Sciences 116, 15849 (2019).DOI
  23. [23]S. Mei and A. Montanari, The generalization error of random features regression: Precise asymptotics and the double descent curve, Communications on Pure and Applied Mathematics 75, 667 (2022).
  24. [24]R. Vershynin, High-dimensional probability: An introduction with applications in data science, Vol. 47 (Cambridge university press, 2018).
  25. [25]V. A. Marčenko and L. A. Pastur, Distribution of eigenvalues for some sets of random matrices, Mathematics of the USSR-Sbornik 1, 457 (1967).DOI
  26. [26]Z. Bai and J. W. Silverstein, Spectral analysis of large dimensional random matrices, Vol. 20 (Springer, 2010).DOI
  27. [27]L. Innocenti, S. Lorenzo, I. Palmisano, F. Albarelli, A. Ferraro, M. Paternostro, and G. M. Palma, Shadow tomography on general measurement frames, PRX Quantum 4, 040328 (2023).DOI
  28. [28]S. Thanasip, S. Wang, M. Cerezo, and Z. Holmes, Exponential concentration in quantum kernel methods, Nature communications 15, 5200 (2024).
  29. [29]M. Kempkes, A. Jia, E. Gil-Fuster, C. Bravo-Prieto, J. Spiegelberg, E. van Nieuwenburg, and V. Dunjko, Double descent in quantum kernel methods, PRX Quantum 7, 010312 (2026).DOI
  30. [30]K. Kamisoyama, L. Nagano, and K. Terashi, Double descent in quantum kernel ridge regression (2026), arXiv:2604.17202 [quant-ph].arxiv.org/abs/2604.17202
  31. [31]V. Belis, J. Bowles, R. Gupta, E. Peters, and M. Schuld, Spectral methods: crucial for machine learning, natural for quantum computers?, arXiv preprint arXiv:2603.24654 (2026).arxiv.org/abs/2603.24654
  32. [32]S. Meyer, E. Yilmaz, A. Lucchi, F. Tacchino, and F. Scala, Shaping double descent in quantum extreme learning through model design, arXiv preprint arXiv:2610.06581 (2026).arxiv.org/abs/2610.06581
  33. [33]A. J. Scott, Tight informationally complete quantum measurements, Journal of Physics A: Mathematical and General 39, 13507 (2006), arXiv:quant-ph/0604049.
  34. [34]F. Mezzadri, How to generate random matrices from the classical compact groups, Notices of the American Mathematical Society 54, 592 (2007).
  35. [35]C. M. Bishop, Training with noise is equivalent to tikhonov regularization, Neural Computation 7, 108 (1995). W. A. Fuller, Measurement Error Models, 1st ed., Wiley Series in Probability and Statistics (Wiley, 1987).DOI
  36. [37]R. J. Carroll, D. Ruppert, L. A. Stefanski, and C. M. Crainiceanu, Measurement Error in Nonlinear Models, 0th ed. (Chapman and Hall/CRC, 2006).
  37. [38]A. A. Mele, Introduction to haar measure tools in quantum information: A beginner’s tutorial, Quantum 8, 1340 (2024).DOI
  38. [39]B. Collins, Moments and cumulants of polynomial random variables on unitary groups, the itzykson-zuber integral, and free probability, International Mathematics Research Notices 2003, 953 (2003).DOI
  39. [40]T. Tao, Topics in random matrix theory, Vol. 132 (American Mathematical Society, 2023).
  40. [41]E. Meckes and M. Meckes, Spectral measures of powers of random matrices, Electronic Communications in Probability 18, 1 (2013).DOI
  41. [42]J. A. Tropp, User-friendly tail bounds for sums of random matrices, Foundations of computational mathematics 12, 389 (2012).
  42. [43]W. Haechem, P. Loubaton, J. Najim, and P. Vallet, On bilinear forms based on the resolvent of large random matrices, Annales de l’Institut Henri Poincaré, Probabilités et Statistiques 49, 36 (2013), arXiv:1004.3848.
  43. [44]W. K. Wootters and B. D. Fields, Optimal state-determination by mutually unbiased measurements, Annals of Physics 191, 363 (1989).DOI
  44. [45]P. Yaskov, A short proof of the marchenko–pastur theorem, Comptes Rendus Mathematique 354, 319 (2016).DOI
  45. [46]A. Carbery and J. Wright, Distributional and l^q norm inequalities for polynomials over convex bodies in R^n, Mathematical research letters 8, 233 (2001).DOI
  46. [47]K. Tikhomirov, Sample covariance matrices of heavy-tailed distributions (2016), arXiv:1606.03557 [math.PR].arxiv.org/abs/1606.03557

Paper details

vlnc:W_zsu6jdcw
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents