We outline refined versions of two major quantum algorithms for performing principal component analysis and solving linear equations. Our methods are exponentially faster than their classical counterparts and even previous quantum algorithms/dequantization algorithms. Oracle/black-box access to classical data is not required, thus implying great capacity for near-term realization. Several applications and implications of these results are discussed. First, we show that a Hamiltonian $H$ with classically known rows/columns can be efficiently simulated, adding another model in addition to the well-known sparse access and linear combination of unitaries models. Second, we provide a simpler proof of the known result that quantum matrix inversion cannot achieve sublinear complexity $κ^{1-γ}$ where $κ$ is the conditional number of the inverted matrix.
We outline refined versions of two major quantum algorithms for performing principal component analysis and solving linear equations. Our methods are exponentially faster than their classical counterparts and even previous quantum algorithms/dequantization algorithms. Oracle/black-box access to classical data is not required, thus implying great capacity for near-term realization. Several applications and implications of these results are discussed. First, we show that a Hamiltonian H with classically known rows/columns can be efficiently simulated, adding another model in addition to the well-known sparse access and linear combination of unitaries models. Second, we provide a simpler proof of the known result that quantum matrix inversion cannot achieve sublinear complexity κ1−γ where κ is the conditional number of the inverted matrix.
Along with the development of quantum technology, finding useful applications of quantum computers has been a central pursuit. Despite various quantum algorithms have been developed, many of them often require strong input assumptions, which is hardware demanding. In particular, recent advances on dequantization have revealed that the quantum advantage is more of a mere artifact of strong input assumptions. In this work, we propose a variant of these algorithms, leveraging both classical and quantum resources. Provided the classical knowledge (the entries) of the matrix/vector of interest, a classical procedure is used to pre-process this information. Then they are fed into a quantum circuit which is shown to be a block encoding of the matrix of interest. From this block-encoding, we show how to use it to tackle a wide range of problems, including principal component analysis, linear equation solving, Hamiltonian simulation, preparing ground state, and data fitting. We also analyze our protocol, showing that both the classical and quantum procedure can achieve logarithmic complexity in the input dimension, thus implying its potential for near term realization. We then discuss several implications and corollaries of our result. First,, our results suggest there are certain matrices/Hamiltonians where our method can provide exponential improvement compared to the existing ones with respect to the sparsity. Regarding dense linear systems, our method achieves exponential speed-up with respect to the inverse of error tolerance, compared to the best previously known quantum algorithm for dense systems. Last, and most importantly, regarding quantum data fitting, we show how the output of our quantum algorithms can be leveraged to predict unseen data. Thus, it provides an end-to-end application, which has been an open aspect of the previous quantum data fitting algorithm.
Abstract Principal component analysis (PCA) is an important dimensionality reduction method in machine learning and data analysis. Recently, the quantum version of PCA has been established to diagonalize quantum states. Although these quantum algorithms promise quantum advantages, they require substantial resources beyond the reach of state-of-the-art quantum technologies. This work aims to reduce resource requirements and improve the efficiency of quantum PCA. Assuming that the quantum state is accessed through a purified quantum query model and a sampling model, we propose quantum algorithms that use minimal resource requirements for ancillary qubits to reveal properties of eigenvectors and eigenvalues of a state. In particular, we show that estimating eigenvalue λ with error ε and success probability larger than λ ( 1 − η ) requests a query complexity O ~ ( ϵ − 1 ) and a sample complexity O ~ ( ϵ − 2 η − 1 ) , respectively. To our knowledge, our result is the first quantum speedup that achieves asymptotic linear scaling in 1 / ϵ for quantum PCA. As applications, we discuss estimating the minimum relative entropy of entanglement of bipartite pure-states and performing quantum state discrimination tasks. We show that quantum speedups are maintained when the pure state has a low Schmidt number and states of discrimination have a low rank. This study opens up a new quantum PCA method for high-dimensional quantum data analysis and discusses its application in quantum information processing tasks.
We propose fast and practical quantum-inspired classical algorithms for solving linear systems. Specifically, given sampling and query access to a matrix A∈Rm×n and a vector b∈Rm, we propose classical algorithms that produce a data structure for the solution x∈Rn of the linear system Ax=b with the ability to sample and query its entries. The resulting x satisfies ∥x−A+b∥≤ε∥A+b∥, where ∥⋅∥ is the spectral norm and A+ is the Moore-Penrose inverse of A. Our algorithm has time complexity O(κF4/κε2) in the general case, where κF=∥A∥F∥A+∥ and κ=∥A∥∥A+∥ are condition numbers. Compared to the prior state-of-the-art result [Shao and Montanaro, arXiv:2103.10309v2], our algorithm achieves a polynomial speedup in condition numbers. When A is s-sparse, our algorithm has complexity O(sκlog(1/ε)), matching the quantum lower bound for solving linear systems in κ and 1/ε up to poly-logarithmic factors [Harrow and Kothari]. When A is s-sparse and symmetric positive-definite, our algorithm has complexity O(sκlog(1/ε)). Technically, our main contribution is the application of the heavy ball momentum method to quantum-inspired classical algorithms for solving linear systems, where we propose two new methods with speedups: quantum-inspired Kaczmarz method with momentum and quantum-inspired coordinate descent method with momentum. Their analysis exploits careful decomposition of the momentum transition matrix and the application of novel spectral norm concentration bounds for independent random matrices. Finally, we also conduct numerical experiments for our algorithms on both synthetic and real-world datasets, and the experimental results support our theoretical claims.
We establish an improved classical algorithm for solving linear systems in a model analogous to the QRAM that is used by quantum linear solvers. Precisely, for the linear system Ax=b , we show that there is a classical algorithm that outputs a data structure for x allowing sampling and querying to the entries, where x is such that ∥x−A+b∥≤ϵ∥A+b∥ . This output can be viewed as a classical analogue to the output of quantum linear solvers. The complexity of our algorithm is O(κF6κ2/ϵ2) , where κF=∥A∥F∥A+∥ and κ=∥A∥∥A+∥ . This improves the previous best algorithm [Gilyén, Song and Tang, arXiv:2009.07268] of complexity O(κF6κ6/ϵ4) . Our algorithm is based on the randomized Kaczmarz method, which is a particular case of stochastic gradient descent. We also find that when A is row sparse, this method already returns an approximate solution x in time O(κF2) , while the best quantum algorithm known returns ∣x⟩ in time O(κF) when A is stored in the QRAM data structure. As a result, assuming access to QRAM and if A is row sparse, the speedup based on current quantum algorithms is quadratic.
Solving systems of linear equations is a key subroutine in many quantum algorithms. In the last 15 years, many quantum linear solvers (QLS) have been developed, competing to achieve the best asymptotic worst-case complexity. Most QLS assume fault-tolerant quantum computers, so they cannot yet be benchmarked on real hardware. Because an algorithm with better asymptotic scaling can underperform on instances of practical interest, the question of which of these algorithms is the most promising remains open. In this work, we implement a method to partially address this question. We consider four well-known QLS algorithms which directly implement an approximate matrix inversion function: the Harrow-Hassidim-Lloyd algorithm, two algorithms utilizing a linear combination of unitaries, and one utilizing the quantum singular value transformation (QSVT). These methods, known as functional QLS, share nearly identical assumptions about the problem setup and oracle access. Their computational cost is dominated by query calls to a matrix oracle encoding the problem one wants to solve. We provide formulas to count the number of queries needed to solve specific problem instances; these can be used to benchmark the algorithms on real-life instances, under benevolent, idealized assumptions, without requiring access to quantum hardware to implement the procedure. We select three data sets: random generated instances that obey the assumptions of functional QLS, linear systems from simplex iterations on MIPLIB, and Poisson equations. Our methods can be easily extended to other data sets and provide a high-level guide to evaluate the performance of a QLS algorithm. In particular, our work shows that HHL underperforms in comparison to the other methods across all data sets, often by orders of magnitude, while the Chebyshev-approximation based methods typically have the best performance.
Quantum algorithms for linear algebra that access a matrix A through quantum random access memory (QRAM) prepare quantum states proportional to the solutions of linear systems and related problems. Dequantized classical algorithms, when given sampling and query access to A, return sampling and query access to an approximate solution. Both kinds of algorithms use a number of queries that is polynomial in the normalized Frobenius norm ρ=∥A∥F/∥A∥ and does not directly depend on the dimension. For linear systems at fixed condition number and accuracy, quantum algorithms use O(ρ) queries, whereas the best dequantized algorithms use O(ρ4). We prove that this quartic gap is tight in the worst case, up to a logarithmic factor. For every constant κ0>1, we construct dense linear systems with condition number at most κ0 for which sampling from the solution within a sufficiently small constant total variation distance requires Ω(ρ4/logρ) classical queries but only Θ(ρ) quantum queries. The same separation holds, for errors below smaller thresholds, for positive-definite systems, least squares, ridge regression, matrix powers and square roots, imaginary-time and real-time Hamiltonian evolution, and sampling from the top principal component. Our results therefore show that many quantum algorithms for linear algebra, achieve a near-quartic speedup in the worst case, over classical algorithms with sampling and query access. All of these classical lower bounds follow from the same framework, which is based on a lower bound for sampling from distributions that depend on a fixed set of Fourier coefficients of an unknown Boolean function.