betaFeedback

WorksPhysics and Astronomy

Improved machine learning algorithm for predicting ground state properties

Abstract

Finding the ground state of a quantum many-body system is a fundamental problem in quantum physics. In this work, we give a classical machine learning (ML) algorithm for predicting ground state properties with an inductive bias encoding geometric locality. The proposed ML model can efficiently predict ground state properties of an nn-qubit gapped local Hamiltonian after learning from only O(log⁡(n))\mathcal{O}(\log(n)) data about other Hamiltonians in the same quantum phase of matter. This improves substantially upon previous results that require O(nc)\mathcal{O}(n^c) data for a large constant cc. Furthermore, the training and prediction time of the proposed ML model scale as O(nlog⁡n)\mathcal{O}(n \log n) in the number of qubits nn. Numerical experiments on physical systems with up to 45 qubits confirm the favorable scaling in predicting ground state properties using a small training dataset.

Introduction

Finding the ground state of a quantum many-body system is a fundamental problem with far-reaching consequences for physics, materials science, and chemistry. Many powerful methods [1, 2, 3, 4, 5, 6, 7] have been proposed, but classical computers still struggle to solve many general classes of the ground state problem. To extend the reach of classical computers, classical machine learning (ML) methods have recently been adapted to study this problem [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28]. A recent work [29] proposes a polynomial-time classical ML algorithm that can efficiently predict ground state properties of gapped geometrically local Hamiltonians, after learning from data obtained by measuring other Hamiltonians in the same quantum phase of matter. Furthermore, [29] shows that under a widely accepted conjecture, no polynomial-time classical algorithm can achieve the same performance guarantee. However, although the ML algorithm given in [29] uses a polynomial amount of training data and computational time, the polynomial scaling O(nc)\mathcal{O}(n^c) has a very large degree cc. Moreover, when the prediction error ϵ\epsilon is small, the amount of training data grows exponentially in 1/ϵ1/\epsilon, indicating that a very small prediction error cannot be achieved efficiently.

In this work, we present an improved ML algorithm for predicting ground state properties. We consider an mm-dimensional vector x∈[−1,1]mx \in[-1,1]^m that parameterizes an nn-qubit gapped geometrically local Hamiltonian given as

H(x)=∑jhj(x⃗j),(1)H(x) = \sum_j h_j(\vec{x}_j), \tag*{(1)}

where xx is the concatenation of constant-dimensional vectors x⃗1,…,x⃗L\vec{x}_1,\ldots,\vec{x}_L parameterizing the few-body interaction hj(x⃗j)h_j(\vec{x}_j). Let ρ(x)\rho(x) be the ground state of H(x)H(x) and OO be a sum of geometrically local observables with ∥O∥∞≤1\lVert O\rVert_\infty\le1. We assume that the geometry of the nn-qubit system is known, but we do not know how hj(x⃗j)h_j(\vec{x}_j) is parameterized or what the observable OO is. The goal is to learn a function h∗(x)h^*(x) that approximates the ground state property Tr⁡(Oρ(x))\operatorname{Tr}(O\rho(x)) from a classical dataset,

(xℓ,yℓ),∀ℓ=1,…,N,(2)(x_\ell,y_\ell), \qquad\forall\ell= 1,\ldots,N, \tag*{(2)}

where yℓ≈Tr⁡(Oρ(xℓ))y_\ell\approx\operatorname{Tr}(O\rho(x_\ell)) records the ground state property for xℓ∈[−1,1]mx_\ell\in[-1,1]^m sampled from an arbitrary unknown distribution D\mathcal{D}.

The setting considered in this work is very similar to that in [29], but we assume the geometry of the nn-qubit system to be known, which is necessary to overcome the sample complexity lower bound of N=nΩ(1/ϵ)N = n^{\Omega(1/\epsilon)} given in [29]. One may compare the setting to that of finding ground states using adiabatic quantum computation [30, 31, 32, 33, 34, 35, 36, 37]. To find the ground state property Tr⁡(Oρ(x))\operatorname{Tr}(O\rho(x)) of H(x)H(x), this class of quantum algorithms requires the ground state ρ0\rho_0 of another Hamiltonian H0H_0 stored in quantum memory, explicit knowledge of a gapped path connecting H0H_0 and H(x)H(x), and an explicit description of OO. In contrast, here we focus on ML algorithms that are entirely classical, have no access to quantum state data, and have no knowledge about the Hamiltonian H(x)H(x), the observable OO, or the gapped paths between H(x)H(x) and other Hamiltonians.

The proposed ML algorithm uses a nonlinear feature map x↦ϕ(x)x \mapsto\phi(x) with a geometric inductive bias built into the mapping. At a high level, the high-dimensional vector ϕ(x)\phi(x) contains nonlinear functions for each geometrically local subset of coordinates in the mm-dimensional vector xx. Here, the geometry over coordinates of the vector xx is defined using the geometry of the nn-qubit system. The ML algorithm learns a function h∗(x)=w∗⋅φ(x)h^*(x) = \mathbf{w}^* \cdot\varphi(x) by training an ℓ1\ell_1-regularized regression (LASSO) [38, 39, 40] in the feature space. We prove that given ϵ=Θ(1)\epsilon= \Theta(1), the improved ML algorithm can use a dataset size of

Overview of the proposed machine learning algorithm

Figure 1. Overview of the proposed machine learning algorithm. Given a vector x∈[−1,1]mx \in[-1,1]^m that parameterizes a quantum many-body Hamiltonian H(x)H(x). The algorithm uses a geometric structure to create a high-dimensional vector φ(x)∈Rmφ\varphi(x) \in\mathbb{R}^{m_\varphi}. The ML algorithm then predicts properties or a representation of the ground state ρ(x)\rho(x) of Hamiltonian H(x)H(x) using the mφm_\varphi-dimensional vector φ(x)\varphi(x).

N=O(log⁡(n)),(3)N = \mathcal{O}(\log(n)), \tag*{(3)}

to learn a function h∗(x)h^*(x) with an average prediction error of at most ϵ\epsilon,

Ex∼D[∣h∗(x)−Tr⁡(Oρ(x))∣2]≤ϵ,(4)\mathbb{E}_{x\sim\mathcal{D}}\left[\left|h^*(x)-\operatorname{Tr}(O\rho(x))\right|^2\right] \leq\epsilon, \tag*{(4)}

with high success probability.

The sample complexity N=O(log⁡(n))N = \mathcal{O}(\log(n)) of the proposed ML algorithm improves substantially over the sample complexity of N=O(nc)N = \mathcal{O}(n^c) in the previously best-known classical ML algorithm [29], where cc is a very large constant. The computational time of both the improved ML algorithm and the ML algorithm in [29] is O(nN)\mathcal{O}(nN). Hence, the logarithmic sample complexity NN immediately implies a nearly linear computational time. In addition to the reduced sample complexity and computational time, the proposed ML algorithm works for any distribution over xx, while the best previously known algorithm [29] works only for the uniform distribution over [−1,1]m[-1,1]^m. Furthermore, when we consider the scaling with the prediction error ϵ\epsilon, the best known classical ML algorithm in [29] has a sample complexity of N=nO(1/ϵ)N = n^{\mathcal{O}(1/\epsilon)}, which is exponential in 1/ϵ1/\epsilon. In contrast, the improved ML algorithm has a sample complexity of N=log⁡(n)2polylog⁡(1/ϵ)N = \log(n)2^{\operatorname{polylog}(1/\epsilon)}, which is quasi-polynomial in 1/ϵ1/\epsilon. In combination with the classical shadow formalism [41, 42, 43, 44, 45], the proposed ML algorithm also yields the same reduction in sample and time complexity compared to [29] for predicting ground state representations.

ML Algorithm and Rigorous Guarantee

The central component of the improved ML algorithm is the geometric inductive bias built into our feature mapping x∈[−1,1]m↦φ(x)∈Rmφx \in[-1,1]^m \mapsto\varphi(x) \in\mathbb{R}^{m_\varphi}. To describe the ML algorithm, we first need to present some definitions relating to this geometric structure.

Definitions

We consider nn qubits arranged at locations, or sites, in a dd-dimensional space, e.g., a spin chain (d=1d = 1), a square lattice (d=2d = 2), or a cubic lattice (d=3d = 3). This geometry is characterized by the distance dqubit(i,i′)d_{\mathrm{qubit}}(i,i') between any two qubits ii and i′i'. Using the distance dqubitd_{\mathrm{qubit}} between qubits, we can define the geometry of local observables. Given any two observables OA,OBO_A,O_B on the nn-qubit system, we define the distance dobs(OA,OB)d_{\mathrm{obs}}(O_A,O_B) between the two observables as the minimum distance between the qubits that OAO_A and OBO_B act on. We also say an observable is geometrically local if it acts nontrivially only on nearby qubits under the distance metric dqubitd_{\mathrm{qubit}}. We then define S(geo)S^{(\mathrm{geo})} as the set of all geometrically local Pauli observables, i.e., geometrically local observables that belong to the set {I,X,Y,Z}⊗n\{I,X,Y,Z\}^{\otimes n}. The size of S(geo)S^{(\mathrm{geo})} is O(n)\mathcal{O}(n), linear in the total number of qubits.

With these basic definitions in place, we now define a few more geometric objects. The first object is the set of coordinates in the mm-dimensional vector xx that are close to a geometrically local Pauli observable PP. This is formally given by,

IP≜{c∈{1,…,m}:dobs(hj(c),P)≤δ1},(5)I_P \triangleq\{c \in\{1,\ldots,m\}:d_{\mathrm{obs}}(h_{j(c)},P)\leq\delta_1\}, \tag*{(5)}

where hj(c)h_{j(c)} is the few-body interaction term in the nn-qubit Hamiltonian H(x)H(x) that is parameterized by the variable xc∈[−1,1]x_c\in[-1,1], and δ1\delta_1 is an efficiently computable hyperparameter that is determined later. Note that, by definition, each variable xcx_c parameterizes one of the interaction terms hj(c)h_{j(c)}. Intuitively, IPI_P is the set of coordinates that have the strongest influence on the function Tr⁡(Pρ(x))\operatorname{Tr}(P\rho(x)).

The second geometric object is a discrete lattice over the space [−1,1]m[-1,1]^m associated to each subset IPI_P of coordinates. For any geometrically local Pauli observable P∈S(geo)P\in S^{(\mathrm{geo})}, we define XPX_P to contain all vectors xx that take on value 0 for coordinates outside IPI_P and take on a set of discrete values for coordinates inside IPI_P. Formally, this is given by

XP≜{x∈[−1,1]m:if c∉IP, xc=0if c∈IP, xc∈{0,±δ2,±2δ2,…,±1}},(6)X_P \triangleq\left\{x\in[-1,1]^m: \begin{array}{ll} \text{if }c\notin I_P,\ x_c=0\\ \text{if }c\in I_P,\ x_c\in\{0,\pm\delta_2,\pm2\delta_2,\ldots,\pm1\} \end{array} \right\}, \tag*{(6)}

where δ2\delta_2 is an efficiently computable hyperparameter to be determined later. The definition of XPX_P is meant to enumerate all sufficiently different vectors for coordinates in the subset IP⊆{1,…,m}I_P\subseteq\{1,\ldots,m\}.

Now given a geometrically local Pauli observable PP and a vector xx in the discrete lattice XP⊆[−1,1]mX_P\subseteq[-1,1]^m, the third object is a set Tx,PT_{x,P} of vectors in [−1,1]m[-1,1]^m that are close to xx for coordinates in IPI_P. This is formally defined as,

Tx,P≜{x′∈[−1,1]m:−δ22<xc−xc′≤δ22, ∀c∈IP}.(7)T_{x,P}\triangleq\left\{x'\in[-1,1]^m:-\frac{\delta_2}{2}<x_c-x'_c\leq\frac{\delta_2}{2},\ \forall c\in I_P\right\}. \tag*{(7)}

The set Tx,PT_{x,P} is defined as a thickened affine subspace close to the vector xx for coordinates in IPI_P. If a vector x′x' is in Tx,PT_{x,P}, then x′x' is close to xx for all coordinates in IPI_P, but x′x' may be far away from xx for coordinates outside of IPI_P.

Feature mapping and ML model

References

  1. [1]P. Hohenberg and W. Kohn. Inhomogeneous electron gas. Phys. Rev., 136:B864–B871, 1964.
  2. [2]W. Kohn. Nobel lecture: Electronic structure of matter—wave functions and density functionals. Rev. Mod. Phys., 71:1253–1266, 1999.DOI
  3. [3]David Ceperley and Berni Alder. Quantum Monte Carlo. Science, 231(4738):555–560, 1986.pubmed.ncbi.nlm.nih.gov/17750966
  4. [4]Anders W. Sandvik. Stochastic series expansion method with operator-loop update. Phys. Rev. B, 59:R14157–R14160, 1999.DOI
  5. [5]Federico Becca and Sandro Sorella. Quantum Monte Carlo Approaches for Correlated Systems. Cambridge University Press, 2017.DOI
  6. [6]Steven R. White. Density matrix formulation for quantum renormalization groups. Phys. Rev. Lett., 69:2863–2866, 1992.pubmed.ncbi.nlm.nih.gov/10046608
  7. [7]Steven R. White. Density-matrix algorithms for quantum renormalization groups. Phys. Rev. B, 48:10345–10356, 1993.pubmed.ncbi.nlm.nih.gov/10007313
  8. [8]Giuseppe Carleo, Ignacio Cirac, Kyle Cranmer, Laurent Daudet, Maria Schuld, Naftali Tishby, Leslie Vogt-Maranto, and Lenka Zdeborová. Machine learning and the physical sciences. Rev. Mod. Phys., 91:045002, 2019.DOI
  9. [9]Juan Carrasquilla. Machine learning for quantum matter. Adv. Phys.: X, 5(1):1797528, 2020.DOI
  10. [10]Dong-Ling Deng, Xiaopeng Li, and S. Das Sarma. Machine learning topological states. Phys. Rev. B, 96:195145, 2017.DOI
  11. [11]Juan Carrasquilla and Roger G. Melko. Machine learning phases of matter. Nat. Phys., 13:431, 2017.
  12. [12]Giuseppe Carleo and Matthias Troyer. Solving the quantum many-body problem with artificial neural networks. Science, 355(6325):602–606, 2017.DOI
  13. [13]Giacomo Torlai and Roger G. Melko. Learning thermodynamics with Boltzmann machines. Physical Review B, 94(16):165134, 2016.
  14. [14]Yusuke Nomura, Andrew S. Darmawan, Youhei Yamaji, and Masatoshi Imada. Restricted boltzmann machine learning for solving strongly correlated quantum systems. Phys. Rev. B, 96:205152, 2017.DOI
  15. [15]Evert P. L. van Nieuwenburg, Ye-Hua Liu, and Sebastian D. Huber. Learning phase transitions by confusion. Nat. Phys., 13:435, 2017.DOI
  16. [16]Lei Wang. Discovering phase transitions with unsupervised learning. Phys. Rev. B, 94:195105, 2016.DOI
  17. [17]Justin Gilmer, Samuel S. Schoenholz, Patrick F Riley, Oriol Vinyals, and George E Dahl. Neural message passing for quantum chemistry. arXiv preprint arXiv:1704.01212, 2017.arxiv.org/abs/1704.01212
  18. [18]Giacomo Torlai, Guglielmo Mazzola, Juan Carrasquilla, Matthias Troyer, Roger Melko, and Giuseppe Carleo. Neural-network quantum state tomography. Nat. Phys., 14(5):447–450, 2018.DOI
  19. [19]Rodrigo A Vargas-Hernández, John Sous, Mona Berciu, and Roman V Krems. Extrapolating quantum observables with machine learning: inferring multiple phase transitions from properties of a single phase. Physical review letters, 121(25):255702, 2018.
  20. [20]KT Schütt, Michael Gastegger, Alexandre Tkatchenko, K-R Müller, and Reinhard J Maurer. Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions. Nat. Commun., 10(1):1–10, 2019.
  21. [21]Ivan Glasser, Nicola Pancotti, Moritz August, Ivan D. Rodriguez, and J. Ignacio Cirac. Neural-network quantum states, string-bond states, and chiral topological states. Phys. Rev. X, 8:011006, 2018.DOI
  22. [22]Matthias C Caro, Hsin-Yuan Huang, Nicholas Ezzell, Joe Gibbs, Andrew T Sornborger, Lukasz Cincio, Patrick J Coles, and Zoë Holmes. Out-of-distribution generalization for learning quantum dynamics. arXiv preprint arXiv:2204.10268, 2022.DOI
  23. [23]Joaquin F Rodriguez-Nieva and Matthias S Scheurer. Identifying topological order through unsupervised machine learning. Nat. Phys., 15(8):790–795, 2019.
  24. [24]Zhuoran Qiao, Matthew Welborn, Animashree Anandkumar, Frederick R Manby, and Thomas F Miller III. OrbNet: Deep learning for quantum chemistry using symmetry-adapted atomic-orbital features. J. Chem. Phys., 153(12):124111, 2020.
  25. [25]Kenny Choo, Antonio Mezzacapo, and Giuseppe Carleo. Fermionic neural-network states for ab-initio electronic structure. Nat. Commun., 11(1):2368, May 2020.DOI
  26. [26]Hiroki Kawai and Yuya O Nakagawa. Predicting excited states from ground state wavefunction by supervised quantum machine learning. Machine Learning: Science and Technology, 1(4):045027, 2020.DOI
  27. [27]Javier Robledo Moreno, Giuseppe Carleo, and Antoine Georges. Deep learning the hohenberg-kohn maps of density functional theory. Physical Review Letters, 125(7):076402, 2020.DOI
  28. [28]Korbinian Kottmann, Philippe Corboz, Maciej Lewenstein, and Antonio Acín. Unsupervised mapping of phase diagrams of 2d systems from infinite projected entangled-pair states via deep anomaly detection. SciPost Physics, 11(2):025, 2021.DOI
  29. [29]Hsin-Yuan Huang, Richard Kueng, Giacomo Torlai, Victor V Albert, and John Preskill. Provably efficient machine learning for quantum many-body problems. arXiv preprint arXiv:2106.12627, 2021.DOI
  30. [30]Edward Farhi, Jeffrey Goldstone, Sam Gutmann, and Michael Sipser. Quantum computation by adiabatic evolution. arXiv preprint quant-ph/0001106, 2000.DOI
  31. [31]Ari Mizel, Daniel A Lidar, and Wim van Dam. Simple proof of equivalence between adiabatic quantum computation and the circuit model. Physical review letters, 99(7):070502, 2007.pubmed.ncbi.nlm.nih.gov/17930879
  32. [32]Andrew M Childs, Edward Farhi, and John Preskill. Robustness of adiabatic quantum computation. Physical Review A, 65(1):012322, 2001.DOI
  33. [33]Dorit Aharonov, Wim Van Dam, Julia Kempe, Zeph Landau, Seth Lloyd, and Oded Regev. Adiabatic quantum computation is equivalent to standard quantum computation. SIAM review, 50(4):755–787, 2008.
  34. [34]Rami Barends, Alireza Shabani, Lucas Lamata, Julian Kelly, Antonio Mezzacapo, U Las Heras, Ryan Babbush, Austin G Fowler, Brooks Campbell, Yu Chen, et al. Digitized adiabatic quantum computing with a superconducting circuit. Nature, 534(7606):222–226, 2016.DOI
  35. [35]Tameem Albash and Daniel A Lidar. Adiabatic quantum computation. Reviews of Modern Physics, 90(1):015002, 2018.DOI
  36. [36]Jiangfeng Du, Nanyang Xu, Xinhua Peng, Pengfei Wang, Sanfeng Wu, and Dawei Lu. Nmr implementation of a molecular hydrogen quantum simulation with adiabatic state preparation. Physical review letters, 104(3):030502, 2010.pubmed.ncbi.nlm.nih.gov/20366636
  37. [37]Kianna Wan and Isaac Kim. Fast digital methods for adiabatic state preparation. arXiv preprint arXiv:2004.04164, 2020.arxiv.org/abs/2004.04164
  38. [38]Fadil Santosa and William W. Symes. Linear inversion of band-limited reflection seismograms. SIAM Journal on Scientific and Statistical Computing, 7(4):1307–1330, 1986.DOI
  39. [39]Robert Tibshirani. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1):267–288, 1996.
  40. [40]Mehryar Mohri, Afshin Rostamizadeh, and Ameet Talwalkar. Foundations of machine learning. The MIT Press, 2018.DOI
  41. [41]Hsin-Yuan Huang, Richard Kueng, and John Preskill. Predicting many properties of a quantum system from very few measurements. Nat. Phys., 16:1050–1057, 2020.DOI
  42. [42]Andreas Elben, Richard Kueng, Hsin-Yuan Huang, Rick van Bijnen, Christian Kokail, Marcello Dalmonte, Pasquale Calabrese, Barbara Kraus, John Preskill, Peter Zoller, and Benoît Vermersch. Mixed-state entanglement from local randomized measurements. Phys. Rev. Lett., 125:200501, 2020.
  43. [43]Andreas Elben, Steven T Flammia, Hsin-Yuan Huang, Richard Kueng, John Preskill, Benoît Vermersch, and Peter Zollner. The randomized measurement toolbox. arXiv preprint arXiv:2201.11374, 2022.
  44. [44]Kianna Wan, William J Huggins, Joonho Lee, and Ryan Babbush. Matchgate shadows for fermionic quantum simulation. arXiv preprint arXiv:2207.13723, 2022.DOI
  45. [45]Kaifeng Bu, Dax Enshan Koh, Roy J Garcia, and Arthur Jaffe. Classical shadows with pauli-invariant unitary ensembles. arXiv preprint arXiv:2202.03272, 2022.DOI
  46. [46]Bradley Efron, Trevor Hastie, Iain Johnstone, and Robert Tibshirani. Least angle regression. The Annals of statistics, 32(2):407–499, 2004.
  47. [47]Ingrid Daubechies, Michel Defrise, and Christine De Mol. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint. Communications on Pure and Applied Mathematics: A Journal Issued by the Courant Institute of Mathematical Sciences, 57(11):1413–1457, 2004.arxiv.org/abs/math/0307152
  48. [48]Patrick L Combettes and Valérie R Wajs. Signal recovery by proximal forward-backward splitting. Multiscale modeling & simulation, 4(4):1168–1200, 2005.DOI
  49. [49]Nicolo Cesa-Bianchi, Shai Shalev-Shwartz, and Ohad Shamir. Efficient learning with partially observed attributes. Journal of Machine Learning Research, 12(10), 2011.arxiv.org/abs/1004.4421
  50. [50]Jerome Friedman, Trevor Hastie, and Rob Tibshirani. Regularization paths for generalized linear models via coordinate descent. Journal of statistical software, 33(1):1, 2010.pubmed.ncbi.nlm.nih.gov/20808728
  51. [51]Elad Hazan and Tomer Koren. Linear regression with limited observation. arXiv preprint arXiv:1206.4678, 2012.arxiv.org/abs/1206.4678
  52. [52]Yanlin Chen and Ronald de Wolf. Quantum algorithms and lower bounds for linear regression with norm constraints. arXiv preprint arXiv:2110.13086, 2021.
  53. [53]Katherine Van Kirk, Jordan Cotler, Hsin-Yuan Huang, and Mikhail D Lukin. Hardware-efficient learning of quantum many-body states. arXiv preprint arXiv:2212.06084, 2022.arxiv.org/abs/2212.06084
  54. [54]Hsin-Yuan Huang, Michael Broughton, Masoud Mohseni, Ryan Babbush, Sergio Boixo, Hartmut Neven, and Jarrod R McClean. Power of data in quantum machine learning. Nat. Commun., 12(1):1–9, 2021.
  55. [55]Sven Bachmann, Spyridon Michalakis, Bruno Nachtergaele, and Robert Sims. Automorphic equivalence within gapped phases of quantum lattice systems. Commun. Math. Phys., 309(3):835–871, 2012.DOI
  56. [56]Matthew B Hastings and Xiao-Gang Wen. Quasidiabatic continuation of quantum states: The stability of topological ground-state degeneracy and emergent gauge invariance. Phys. Rev. B, 72(4):045141, 2005.DOI
  57. [57]Tobias J Osborne. Simulating adiabatic evolution of gapped spin systems. Phys. Rev. A, 75(3):032321, 2007.DOI
  58. [58]Hsin-Yuan Huang, Sitan Chen, and John Preskill. Learning to predict arbitrary quantum processes. arXiv preprint arXiv:2210.14894, 2022.arxiv.org/abs/2210.14894
  59. [59]Ali Rahimi and Benjamin Recht. Random features for large-scale kernel machines. Advances in neural information processing systems, 20, 2007.
  60. [60]Steven R White. Density matrix formulation for quantum renormalization groups. Physical review letters, 69(19):2863, 1992.pubmed.ncbi.nlm.nih.gov/10046608
  61. [61]Ulrich Schollwœck. The density-matrix renormalization group in the age of matrix product states. Ann. Phys., 326(1):96 – 192, 2011. January 2011 Special Issue.DOI
  62. [62]Corinna Cortes and Vladimir Vapnik. Support-vector networks. Mach. Learn., 20(3):273–297, 1995.DOI
  63. [63]Kevin P Murphy. Machine learning: a probabilistic perspective. MIT Press, 2012.
  64. [64]Arthur Jacot, Franck Gabriel, and Clément Hongler. Neural tangent kernel: Convergence and generalization in neural networks. In NeurIPS, pages 8571–8580, 2018.arxiv.org/abs/1806.07572
  65. [65]Roman Novak, Lechao Xiao, Jakeir Hu, Jiri Hron, Jaehoon Lee, Alexander A. Alemi, Jascha Sohl-Dickstein, and Samuel S. Schoenholz. Neural tangents: Fast and easy infinite neural networks in python. In International Conference on Learning Representations, 2020.arxiv.org/abs/1912.02803
  66. [66]Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901, 2020.arxiv.org/abs/2005.14165
  67. [67]Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009.
  68. [68]Chitwan Saharia, William Chan, Saurabh Saxena, Lala Li, Jay Whang, Emily Denton, Seyed Kamyar Seyed Ghasemipour, Burcu Karagol Ayan, S Sara Mahdavi, Rapha Gontijo Lopes, et al. Photorealistic text-to-image diffusion models with deep language understanding. arXiv preprint arXiv:2205.11487, 2022.arxiv.org/abs/2205.11487
  69. [69]Haoxiang Wang, Maurice Weber, Josh Isaac, and Cedric Yen-Yu Lin. Predicting properties of quantum systems with conditional generative models. arXiv preprint arXiv:2211.16943, 2022.arxiv.org/abs/2211.16943
  70. [70]Viet T Tran, Laura Lewis, Hsin-Yuan Huang, Johannes Kofler, Richard Kueng, Sepp Hochreiter, and Sebastian Lehner. Using shadows to learn ground state properties of quantum hamiltonians. Machine Learning and Physical Sciences Workshop at the 36th Conference on Neural Information Processing Systems (NeurIPS), 2022.
  71. [71]Dorit Aharonov, Jordan S Cotler, and Xiao-Liang Qi. Quantum algorithmic measurement. arXiv preprint arXiv:2101.04634, 2021.
  72. [72]Sitan Chen, Jordan Cotler, Hsin-Yuan Huang, and Jerry Li. Exponential separations between learning with and without quantum memory. In 2021 IEEE 62nd Annual Symposium on Foundations of Computer Science (FOCS), pages 574–585. IEEE, 2022.arxiv.org/abs/2111.05881
  73. [73]Hsin-Yuan Huang, Steven T Flammia, and John Preskill. Foundations for learning from noisy quantum experiments. arXiv preprint arXiv:2204.13691, 2022.arxiv.org/abs/2204.13691
  74. [74]Hsin-Yuan Huang, Richard Kueng, and John Preskill. Information-theoretic bounds on quantum advantage in machine learning. Phys. Rev. Lett., 126:190505, 2021.DOI
  75. [75]Hsin-Yuan Huang, Michael Broughton, Jordan Cotler, Sitan Chen, Jerry Li, Masoud Mohseni, Hartmut Neven, Ryan Bab-bush, Richard Kueng, John Preskill, et al. Quantum advantage in learning from experiments. Science, 376(6598):1182–1186, 2022.DOI
  76. [76]Emilio Onorati, Cambyse Rouzé, Daniel Stilck França, and James D. Watson. Efficient learning of lattice quantum systems and phases of matter. to appear on arXiv, 2023.
  77. [77]Elliott H Lieb and Derek W Robinson. The finite group velocity of quantum spin systems. In Statistical mechanics, pages 425–431. Springer, 1972.
  78. [78]Matthew B Hastings. Locality in quantum systems. arXiv:1008.5137, 2010.arxiv.org/abs/1008.5137
  79. [79]Sergey Bravyi, Matthew B Hastings, and Frank Verstraete. Lieb-robinson bounds and the generation of correlations and topological quantum order. Phys. Rev. Lett., 97(5):050401, 2006.pubmed.ncbi.nlm.nih.gov/17026080
  80. [80]Iosif Pinelis. Exact lower and upper bounds on the incomplete gamma function. arXiv preprint arXiv:2005.06384, 2020.arxiv.org/abs/2005.06384
  81. [81]Andrew J. Ferris and Guifre Vidal. Perfect sampling with unitary tensor networks. Phys. Rev. B, 85:165146, 2012.DOI
  82. [82]F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res., 12:2825–2830, 2011.arxiv.org/abs/1201.0490

Paper details

vlnc:W_mx2sq2eq
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