Learning with Optimized Random Features: Exponential Speedup by Quantum Machine Learning without Sparsity and Low-Rank Assumptions

Kernel methods augmented with random features give scalable algorithms for learning from big data. But it has been computationally hard to sample random features according to a probability distribution that is optimized for the data, so as to minimize the required number of features for achieving the learning to a desired accuracy. Here, we develop a quantum algorithm for sampling from this optimized distribution over features, in runtime $O(D)$ that is linear in the dimension $D$ of the input data. Our algorithm achieves an exponential speedup in $D$ compared to any known classical algorithm for this sampling task. In contrast to existing quantum machine learning algorithms, our algorithm circumvents sparsity and low-rank assumptions and thus has wide applicability. We also show that the sampled features can be combined with regression by stochastic gradient descent to achieve the learning without canceling out our exponential speedup. Our algorithm based on sampling optimized random features leads to an accelerated framework for machine learning that takes advantage of quantum computers.

[1]  L.-M. Duan,et al.  Experimental realization of 105-qubit random access quantum memory , 2019, npj Quantum Information.

[2]  Matthias W. Seeger,et al.  Using the Nyström Method to Speed Up Kernel Machines , 2000, NIPS.

[3]  S. Aaronson Read the fine print , 2015, Nature Physics.

[4]  Iordanis Kerenidis,et al.  Quantum Recommendation Systems , 2016, ITCS.

[5]  Francis R. Bach,et al.  On the Equivalence between Kernel Quadrature Rules and Random Feature Expansions , 2015, J. Mach. Learn. Res..

[6]  Seth Lloyd,et al.  Quantum algorithm for data fitting. , 2012, Physical review letters.

[7]  A. Harrow,et al.  Quantum algorithm for linear systems of equations. , 2008, Physical review letters.

[8]  Richard Cleve,et al.  Fast parallel circuits for the quantum Fourier transform , 2000, Proceedings 41st Annual Symposium on Foundations of Computer Science.

[9]  Richard Jozsa,et al.  Implementing smooth functions of a Hermitian matrix on a quantum computer , 2018, Journal of Physics Communications.

[10]  Andris Ambainis,et al.  Variable time amplitude amplification and quantum algorithms for linear algebra problems , 2012, STACS.

[11]  Prateek Jain,et al.  Making the Last Iterate of SGD Information Theoretically Optimal , 2019, COLT.

[12]  Daniele Calandriello,et al.  On Fast Leverage Score Sampling and Optimal Learning , 2018, NeurIPS.

[13]  Katya Scheinberg,et al.  Efficient SVM Training Using Low-Rank Kernel Representations , 2002, J. Mach. Learn. Res..

[14]  Franccois Le Gall,et al.  Quantum-Inspired Classical Algorithms for Singular Value Transformation , 2020, MFCS.

[15]  Sean Hallgren,et al.  An improved quantum Fourier transform algorithm and applications , 2000, Proceedings 41st Annual Symposium on Foundations of Computer Science.

[16]  C.E. Shannon,et al.  Communication in the Presence of Noise , 1949, Proceedings of the IRE.

[17]  Yiming Yang,et al.  Data-driven Random Fourier Features using Stein Effect , 2017, IJCAI.

[18]  Seth Lloyd,et al.  Quantum random access memory. , 2007, Physical review letters.

[19]  Jie Yang,et al.  Random Fourier Features via Fast Surrogate Leverage Weighted Sampling , 2019, AAAI.

[20]  Sanjiv Kumar,et al.  Orthogonal Random Features , 2016, NIPS.

[21]  Lov K. Grover,et al.  Creating superpositions that correspond to efficiently integrable probability distributions , 2002, quant-ph/0208112.

[22]  Andrew M. Childs,et al.  Quantum Algorithm for Systems of Linear Equations with Exponentially Improved Dependence on Precision , 2015, SIAM J. Comput..

[23]  Hans-J. Briegel,et al.  Machine learning \& artificial intelligence in the quantum domain , 2017, ArXiv.

[24]  John G. Proakis,et al.  Digital Signal Processing: Principles, Algorithms, and Applications , 1992 .

[25]  Bernhard Schölkopf,et al.  Sparse Greedy Matrix Approximation for Machine Learning , 2000, International Conference on Machine Learning.

[26]  Harvey,et al.  Integer multiplication in time O(n log n) , 2021, Annals of Mathematics.

[27]  Peter W. Shor,et al.  Polynomial-Time Algorithms for Prime Factorization and Discrete Logarithms on a Quantum Computer , 1995, SIAM Rev..

[28]  Vikas Sindhwani,et al.  Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels , 2014, J. Mach. Learn. Res..

[29]  Shahin Shahrampour,et al.  On Sampling Random Features From Empirical Leverage Scores: Implementation and Theoretical Guarantees , 2019, ArXiv.

[30]  Kristan Temme,et al.  Supervised learning with quantum-enhanced feature spaces , 2018, Nature.

[31]  Alexander J. Smola,et al.  Fastfood - Computing Hilbert Space Expansions in loglinear time , 2013, ICML.

[32]  Raman Arora,et al.  Streaming Kernel PCA with \tilde{O}(\sqrt{n}) Random Features , 2018, NeurIPS.

[33]  Tongyang Li,et al.  Sampling-based sublinear low-rank matrix arithmetic framework for dequantizing Quantum machine learning , 2019, STOC.

[34]  Martin Rötteler,et al.  Optimizing Quantum Circuits for Arithmetic , 2018, ArXiv.

[35]  Ashley Montanaro,et al.  Quantum algorithms: an overview , 2015, npj Quantum Information.

[36]  Anthony Widjaja,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2003, IEEE Transactions on Neural Networks.

[37]  Nathan Wiebe,et al.  Quantum singular value transformation and beyond: exponential improvements for quantum matrix arithmetics , 2018, STOC.

[38]  Liang Jiang,et al.  Hardware-Efficient Quantum Random Access Memory with Hybrid Quantum Acoustic Systems. , 2019, Physical review letters.

[39]  Lorenzo Rosasco,et al.  Learning with SGD and Random Features , 2018, NeurIPS.

[40]  Ievgeniia Oshurko Quantum Machine Learning , 2020, Quantum Computing.

[41]  S. Lloyd,et al.  Quantum principal component analysis , 2013, Nature Physics.

[42]  AI Koan,et al.  Weighted Sums of Random Kitchen Sinks: Replacing minimization with randomization in learning , 2008, NIPS.

[43]  S. Lloyd,et al.  Architectures for a quantum random access memory , 2008, 0807.4994.

[44]  Joseph Fitzsimons,et al.  Quantum assisted Gaussian process regression , 2015, Physical Review A.

[45]  Ameya Velingker,et al.  Random Fourier Features for Kernel Ridge Regression: Approximation Bounds and Statistical Guarantees , 2018, ICML.

[46]  Alessandra Di Pierro,et al.  Kernel methods in Quantum Machine Learning , 2019, Quantum Machine Intelligence.

[47]  Le Song,et al.  Scalable Kernel Methods via Doubly Stochastic Gradients , 2014, NIPS.

[48]  John C. Duchi,et al.  Learning Kernels with Random Features , 2016, NIPS.

[49]  Ewin Tang,et al.  A quantum-inspired classical algorithm for recommendation systems , 2018, Electron. Colloquium Comput. Complex..

[50]  Francis R. Bach,et al.  Sharp analysis of low-rank kernel matrix approximations , 2012, COLT.

[51]  Seth Lloyd,et al.  Quantum algorithms for topological and geometric analysis of data , 2016, Nature Communications.

[52]  Simone Severini,et al.  Quantum machine learning: a classical perspective , 2017, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[53]  Ambuj Tewari,et al.  But How Does It Work in Theory? Linear SVM with Random Features , 2018, NeurIPS.

[54]  Isaac L. Chuang,et al.  Quantum Computation and Quantum Information (10th Anniversary edition) , 2011 .

[55]  L. Wossnig,et al.  Quantum Linear System Algorithm for Dense Matrices. , 2017, Physical review letters.

[56]  Lorenzo Rosasco,et al.  Generalization Properties of Learning with Random Features , 2016, NIPS.

[57]  Travis S. Humble,et al.  Quantum supremacy using a programmable superconducting processor , 2019, Nature.

[58]  Nicholas J. A. Harvey,et al.  Tight Analyses for Non-Smooth Stochastic Gradient Descent , 2018, COLT.

[59]  Zhu Li,et al.  Towards a Unified Analysis of Random Fourier Features , 2018, ICML.

[60]  Stacey Jeffery,et al.  The power of block-encoded matrix powers: improved regression techniques via faster Hamiltonian simulation , 2018, ICALP.

[61]  Felipe Cucker,et al.  On the mathematical foundations of learning , 2001 .

[62]  Michael W. Mahoney,et al.  Fast Randomized Kernel Ridge Regression with Statistical Guarantees , 2015, NIPS.

[63]  Benjamin Recht,et al.  Random Features for Large-Scale Kernel Machines , 2007, NIPS.