Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:1505.08126 · Leibniz international proceedings in informatics

A Quasi-Random Approach to Matrix Spectral Analysis

Michael Ben-Or, Lior Eldar

Published May 29, 2015• 4 citations
arXivHTMLPDF

Abstract

Inspired by the quantum computing algorithms for Linear Algebra problems [HHL,TaShma] we study how the simulation on a classical computer of this type of "Phase Estimation algorithms" performs when we apply it to solve the Eigen-Problem of Hermitian matrices. The result is a completely new, efficient and stable, parallel algorithm to compute an approximate spectral decomposition of any Hermitian matrix. The algorithm can be implemented by Boolean circuits in $O(\log^2 n)$ parallel time with a total cost of $O(n^{ω+1})$ Boolean operations. This Boolean complexity matches the best known rigorous $O(\log^2 n)$ parallel time algorithms, but unlike those algorithms our algorithm is (logarithmically) stable, so further improvements may lead to practical implementations. All previous efficient and rigorous approaches to solve the Eigen-Problem use randomization to avoid bad condition as we do too. Our algorithm makes further use of randomization in a completely new way, taking random powers of a unitary matrix to randomize the phases of its eigenvalues. Proving that a tiny Gaussian perturbation and a random polynomial power are sufficient to ensure almost pairwise independence of the phases $(\mod (2π))$ is the main technical contribution of this work. This randomization enables us, given a Hermitian matrix with well separated eigenvalues, to sample a random eigenvalue and produce an approximate eigenvector in $O(\log^2 n)$ parallel time and $O(n^ω)$ Boolean complexity. We conjecture that further improvements of our method can provide a stable solution to the full approximate spectral decomposition problem with complexity similar to the complexity (up to a logarithmic factor) of sampling a single eigenvector.

DOI
10.4230/lipics.itcs.2018.6

Related papers

Computing eigenvalues of diagonalizable matrices in a quantum computer

Changpeng Shao

Solving linear systems and computing eigenvalues are two fundamental problems in linear algebra. For solving linear systems, many efficient quantum algorithms have been discovered. For computing eigenvalues, currently, we have efficient quantum algorithms for Hermitian and unitary matrices. However, the general case is far from fully understood. Combining quantum phase estimation, quantum algorithm to solve linear differential equations and quantum singular value estimation, we propose two quantum algorithms to compute the eigenvalues of diagonalizable matrices that only have real eigenvalues and normal matrices. The output of the quantum algorithms is a superposition of the eigenvalues and the corresponding eigenvectors. The complexities are dominated by solving a linear system of ODEs and performing quantum singular value estimation, which usually can be solved efficiently in a quantum computer. In the special case when the matrix $M$ is $s$-sparse, the complexity is $\widetilde{O}(sρ^2 κ^2/ε^2)$ for diagonalizable matrices that only have real eigenvalues, and $\widetilde{O}(sρ\|M\|_{\max} /ε^2)$ for normal matrices. Here $ρ$ is an upper bound of the eigenvalues, $κ$ is the conditioning of the eigenvalue problem, and $ε$ is the precision to approximate the eigenvalues. We also extend the quantum algorithm to diagonalizable matrices with complex eigenvalues under an extra assumption.

arXiv:1912.08015•Dec 17, 2019•12 citations

Randomized Adiabatic Quantum Linear Solver Algorithm with Optimal Complexity Scaling and Detailed Running Costs

David Jennings, Matteo Lostaglio, Sam Pallister +2 more

Solving linear systems of equations is a fundamental problem with a wide variety of applications across many fields of science, and there is increasing effort to develop quantum linear solver algorithms. Subaşı . [Phys. Rev. Lett. 122, 060504 (2019)] proposed a randomized algorithm inspired by adiabatic quantum computing, based on a sequence of random Hamiltonian simulation steps, with suboptimal scaling in the condition number κ of the linear system and the target error ϵ . Here we go beyond these results in several ways. Firstly, using filtering [Lin and Tong, Quantum 4, 361 (2020)] and Poissonization techniques [Cunningham and Roland, ArXiv:2406.03972 (2024)], the algorithm complexity is improved to the optimal scaling O ( κ log  ( 1 / ϵ ) ) —an exponential improvement in ϵ , and a shaving of a log  κ scaling factor in κ . Secondly, the algorithm is further modified to achieve constant factor improvements, which are vital as we progress towards hardware implementations on fault-tolerant devices. We introduce a cheaper randomized walk operator method replacing Hamiltonian simulation—which also removes the need for potentially challenging classical precomputations; randomized routines are sampled over optimized random variables; circuit constructions are improved. We obtain a closed formula rigorously upper bounding the expected number of times one needs to apply a block-encoding of the linear system matrix to output a quantum state encoding the solution to the linear system. The upper bound is 837 κ at ϵ = 10 − 10 for Hermitian matrices.

arXiv:2305.11352•PRX Quantum•Oct 7, 2025•52 citations

Resolvent-based quantum phase estimation: Towards estimation of parametrized eigenvalues

Abhijeet Alase, Salini Karuvade

Quantum algorithms for estimating the eigenvalues of matrices, including the phase estimation algorithm, serve as core subroutines in a wide range of quantum algorithms, including those in quantum chemistry and quantum machine learning. The standard quantum eigenvalue (phase) estimation algorithm accepts a Hermitian (unitary) matrix and a state in an unknown superposition of its eigenstates as input, and coherently records the estimates for real eigenvalues (eigenphases) in an ancillary register. Extension of quantum eigenvalue and phase estimation algorithms to the case of non-normal input matrices is obstructed by several factors such as non-orthogonality of eigenvectors, existence of generalized eigenvectors and the fact that eigenvalues may lie anywhere on the complex plane. In this work, we propose a novel approach for estimating the eigenvalues of non-normal matrices based on preparation of a state that we call the "resolvent state". We construct the first efficient algorithm for estimating the phases of the unimodular eigenvalues of a given non-unitary matrix. We then construct an efficient algorithm for estimating the real eigenvalues of a given non-Hermitian matrix, achieving complexities that match the best known results while operating under significantly relaxed assumptions on the non-real part of the spectrum. The resolvent-based approach that we introduce also extends to estimating eigenvalues that lie on a parametrized complex curve, subject to explicitly stated conditions, thereby paving the way for a new paradigm of parametric eigenvalue estimation.

arXiv:2410.04837•Oct 7, 2024•7 citations

Approximating Hamiltonian dynamics with the Nyström method

Alessandro Rudi, Leonard Wossnig, Carlo Ciliberto +3 more

Simulating the time-evolution of quantum mechanical systems is BQP-hard and expected to be one of the foremost applications of quantum computers. We consider classical algorithms for the approximation of Hamiltonian dynamics using subsampling methods from randomized numerical linear algebra. We derive a simulation technique whose runtime scales polynomially in the number of qubits and the Frobenius norm of the Hamiltonian. As an immediate application, we show that sample based quantum simulation, a type of evolution where the Hamiltonian is a density matrix, can be efficiently classically simulated under specific structural conditions. Our main technical contribution is a randomized algorithm for approximating Hermitian matrix exponentials. The proof leverages a low-rank, symmetric approximation via the Nyström method. Our results suggest that under strong sampling assumptions there exist classical poly-logarithmic time simulations of quantum computations.

arXiv:1804.02484•Quantum•Feb 20, 2020•12 citations

Quantum Eigensolver for Non-Normal Matrices via Ground State Energy Estimation

Honghong Lin, Yun Shang

Large-scale eigenvalue problems pose a significant challenge to classical computers. While there are efficient quantum algorithms for unitary or Hermitian matrices, eigenvalue problems for non-normal matrices remain open in quantum computing. In this work, we propose a quantum algorithm that given a non-normal matrix, outputs an estimate of an eigenvalue to within additive error $ε$ with probability at least $1-p_{\rm fail}$. Our estimation strategy is to sample points on the complex plane and examine the distance between the sampled point and the eigenvalues. We show that the distance is related to the smallest singular value of the shifted matrix, hence reducing the problem to ground state energy estimation via Hermitianization. With the knowledge of an eigenvalue, we are able to prepare the associated eigenvector using ground state preparation. Our estimating scheme can also be modified to approximate the extreme eigenvalue, and in particular the spectral gap. The algorithm is implemented based on the block encoding input model and requires $O(κ^2ε^{-(2m-1)}\log(1/p_{\rm fail}))$ queries to the block encoding oracle. Our algorithm is the first general eigenvalue algorithm that achieves this scaling. We also perform numerical simulation to validate the algorithms.

arXiv:2502.18119•Feb 25, 2025•3 citations

A Quantum Algorithm for Computing All Eigenvalues and Eigenstates of Sparse Hermitian Matrices

Krishnakanta Barik, Goutam Paul

Computation of eigenvalues and eigenvectors of a sparse Hermitian matrix is a fundamental problem in algebra, arising both as a problem of independent interest and as a key subroutine in a wide range of computational and scientific applications. In this work, we develop a quantum algorithm that determines the spectrum of a sparse Hermitian matrix, defined as the set of its distinct eigenvalues, and for each eigenvalue, prepares a corresponding eigenvector as a quantum state in its eigenspace. The eigenvalues are represented as binary strings with a prescribed precision, while the corresponding eigenvectors are kept as quantum states, eliminating the need to reconstruct their classical counterparts. The algorithm instead enables estimation of expectation values $\braket{x|M|x}$, where $\ket{x}$ is a prepared eigenstate and $M$ is an observable. For an $n \times n$ $s$-sparse Hermitian matrix $H$, dense methods based on Householder tridiagonalization and QR iteration compute all eigenvalues and eigenvectors in $\mathcal{O}(n^3)$ time. Exploiting sparsity, the Lanczos method computes all eigenvalues in $\mathcal{O}(n^2 s)$ time in exact arithmetic, but recovering all eigenvectors still requires $\mathcal{O}(n^3)$ time. In contrast, our algorithm runs in time $\tilde{\mathcal{O}}(n s^2 \log^2 n / Δ_{\min})$, where $Δ_{\min}$ denotes the minimum spectral gap of the normalized matrix. For inverse gaps up to $Δ_{\min}^{-1} = \mathcal{O}(n)$, the most favorable scaling for $n$ distinct eigenvalues, attained by nearly uniformly spaced spectra such as those of perfect-state-transfer spin chains and the Sylvester-Kac matrix, and for sparsity up to $s = o(\sqrt{n}/\log n)$, the proposed algorithm achieves an improved asymptotic runtime over the classical $\mathcal{O}(n^3)$ approach for computing all eigenvalues and eigenvectors, under standard sparse-access assumptions.

arXiv:2611.16610•Nov 12, 2026•0 citations
Inspired by the quantum computing algorithms for Linear Algebra problems [HHL,TaShma] we study how the simulation on a classical computer of this type of "Phase Estimation algorithms" performs when we apply it to solve the Eigen-Problem of Hermitian matrices. The result is a completely new, efficient and stable, parallel algorithm to compute an approximate spectral decomposition of any Hermitian matrix. The algorithm can be implemented by Boolean circuits in O(log⁡2n)O(\log^2 n)O(log2n) parallel time with a total cost of O(nω+1)O(n^{ω+1})O(nω+1) Boolean operations. This Boolean complexity matches the best known rigorous O(log⁡2n)O(\log^2 n)O(log2n) parallel time algorithms, but unlike those algorithms our algorithm is (logarithmically) stable, so further improvements may lead to practical implementations. All previous efficient and rigorous approaches to solve the Eigen-Problem use randomization to avoid bad condition as we do too. Our algorithm makes further use of randomization in a completely new way, taking random powers of a unitary matrix to randomize the phases of its eigenvalues. Proving that a tiny Gaussian perturbation and a random polynomial power are sufficient to ensure almost pairwise independence of the phases (mod  (2π))(\mod (2π))(mod(2π)) is the main technical contribution of this work. This randomization enables us, given a Hermitian matrix with well separated eigenvalues, to sample a random eigenvalue and produce an approximate eigenvector in O(log⁡2n)O(\log^2 n)O(log2n) parallel time and O(nω)O(n^ω)O(nω) Boolean complexity. We conjecture that further improvements of our method can provide a stable solution to the full approximate spectral decomposition problem with complexity similar to the complexity (up to a logarithmic factor) of sampling a single eigenvector.
Solving linear systems and computing eigenvalues are two fundamental problems in linear algebra. For solving linear systems, many efficient quantum algorithms have been discovered. For computing eigenvalues, currently, we have efficient quantum algorithms for Hermitian and unitary matrices. However, the general case is far from fully understood. Combining quantum phase estimation, quantum algorithm to solve linear differential equations and quantum singular value estimation, we propose two quantum algorithms to compute the eigenvalues of diagonalizable matrices that only have real eigenvalues and normal matrices. The output of the quantum algorithms is a superposition of the eigenvalues and the corresponding eigenvectors. The complexities are dominated by solving a linear system of ODEs and performing quantum singular value estimation, which usually can be solved efficiently in a quantum computer. In the special case when the matrix MMM is sss-sparse, the complexity is O~(sρ2κ2/ε2)\widetilde{O}(sρ^2 κ^2/ε^2)O(sρ2κ2/ε2) for diagonalizable matrices that only have real eigenvalues, and O~(sρ∥M∥max⁡/ε2)\widetilde{O}(sρ\|M\|_{\max} /ε^2)O(sρ∥M∥max​/ε2) for normal matrices. Here ρρρ is an upper bound of the eigenvalues, κκκ is the conditioning of the eigenvalue problem, and εεε is the precision to approximate the eigenvalues. We also extend the quantum algorithm to diagonalizable matrices with complex eigenvalues under an extra assumption.
Solving linear systems of equations is a fundamental problem with a wide variety of applications across many fields of science, and there is increasing effort to develop quantum linear solver algorithms. Subaşı . [Phys. Rev. Lett. 122, 060504 (2019)] proposed a randomized algorithm inspired by adiabatic quantum computing, based on a sequence of random Hamiltonian simulation steps, with suboptimal scaling in the condition number κ of the linear system and the target error ϵ . Here we go beyond these results in several ways. Firstly, using filtering [Lin and Tong, Quantum 4, 361 (2020)] and Poissonization techniques [Cunningham and Roland, ArXiv:2406.03972 (2024)], the algorithm complexity is improved to the optimal scaling O ( κ log  ( 1 / ϵ ) ) —an exponential improvement in ϵ , and a shaving of a log  κ scaling factor in κ . Secondly, the algorithm is further modified to achieve constant factor improvements, which are vital as we progress towards hardware implementations on fault-tolerant devices. We introduce a cheaper randomized walk operator method replacing Hamiltonian simulation—which also removes the need for potentially challenging classical precomputations; randomized routines are sampled over optimized random variables; circuit constructions are improved. We obtain a closed formula rigorously upper bounding the expected number of times one needs to apply a block-encoding of the linear system matrix to output a quantum state encoding the solution to the linear system. The upper bound is 837 κ at ϵ = 10 − 10 for Hermitian matrices.
Quantum algorithms for estimating the eigenvalues of matrices, including the phase estimation algorithm, serve as core subroutines in a wide range of quantum algorithms, including those in quantum chemistry and quantum machine learning. The standard quantum eigenvalue (phase) estimation algorithm accepts a Hermitian (unitary) matrix and a state in an unknown superposition of its eigenstates as input, and coherently records the estimates for real eigenvalues (eigenphases) in an ancillary register. Extension of quantum eigenvalue and phase estimation algorithms to the case of non-normal input matrices is obstructed by several factors such as non-orthogonality of eigenvectors, existence of generalized eigenvectors and the fact that eigenvalues may lie anywhere on the complex plane. In this work, we propose a novel approach for estimating the eigenvalues of non-normal matrices based on preparation of a state that we call the "resolvent state". We construct the first efficient algorithm for estimating the phases of the unimodular eigenvalues of a given non-unitary matrix. We then construct an efficient algorithm for estimating the real eigenvalues of a given non-Hermitian matrix, achieving complexities that match the best known results while operating under significantly relaxed assumptions on the non-real part of the spectrum. The resolvent-based approach that we introduce also extends to estimating eigenvalues that lie on a parametrized complex curve, subject to explicitly stated conditions, thereby paving the way for a new paradigm of parametric eigenvalue estimation.
Simulating the time-evolution of quantum mechanical systems is BQP-hard and expected to be one of the foremost applications of quantum computers. We consider classical algorithms for the approximation of Hamiltonian dynamics using subsampling methods from randomized numerical linear algebra. We derive a simulation technique whose runtime scales polynomially in the number of qubits and the Frobenius norm of the Hamiltonian. As an immediate application, we show that sample based quantum simulation, a type of evolution where the Hamiltonian is a density matrix, can be efficiently classically simulated under specific structural conditions. Our main technical contribution is a randomized algorithm for approximating Hermitian matrix exponentials. The proof leverages a low-rank, symmetric approximation via the Nyström method. Our results suggest that under strong sampling assumptions there exist classical poly-logarithmic time simulations of quantum computations.
Large-scale eigenvalue problems pose a significant challenge to classical computers. While there are efficient quantum algorithms for unitary or Hermitian matrices, eigenvalue problems for non-normal matrices remain open in quantum computing. In this work, we propose a quantum algorithm that given a non-normal matrix, outputs an estimate of an eigenvalue to within additive error εεε with probability at least 1−pfail1-p_{\rm fail}1−pfail​. Our estimation strategy is to sample points on the complex plane and examine the distance between the sampled point and the eigenvalues. We show that the distance is related to the smallest singular value of the shifted matrix, hence reducing the problem to ground state energy estimation via Hermitianization. With the knowledge of an eigenvalue, we are able to prepare the associated eigenvector using ground state preparation. Our estimating scheme can also be modified to approximate the extreme eigenvalue, and in particular the spectral gap. The algorithm is implemented based on the block encoding input model and requires O(κ2ε−(2m−1)log⁡(1/pfail))O(κ^2ε^{-(2m-1)}\log(1/p_{\rm fail}))O(κ2ε−(2m−1)log(1/pfail​)) queries to the block encoding oracle. Our algorithm is the first general eigenvalue algorithm that achieves this scaling. We also perform numerical simulation to validate the algorithms.
Computation of eigenvalues and eigenvectors of a sparse Hermitian matrix is a fundamental problem in algebra, arising both as a problem of independent interest and as a key subroutine in a wide range of computational and scientific applications. In this work, we develop a quantum algorithm that determines the spectrum of a sparse Hermitian matrix, defined as the set of its distinct eigenvalues, and for each eigenvalue, prepares a corresponding eigenvector as a quantum state in its eigenspace. The eigenvalues are represented as binary strings with a prescribed precision, while the corresponding eigenvectors are kept as quantum states, eliminating the need to reconstruct their classical counterparts. The algorithm instead enables estimation of expectation values ⟨x∣M∣x⟩\braket{x|M|x}⟨x∣M∣x⟩, where ∣x⟩\ket{x}∣x⟩ is a prepared eigenstate and MMM is an observable. For an n×nn \times nn×n sss-sparse Hermitian matrix HHH, dense methods based on Householder tridiagonalization and QR iteration compute all eigenvalues and eigenvectors in O(n3)\mathcal{O}(n^3)O(n3) time. Exploiting sparsity, the Lanczos method computes all eigenvalues in O(n2s)\mathcal{O}(n^2 s)O(n2s) time in exact arithmetic, but recovering all eigenvectors still requires O(n3)\mathcal{O}(n^3)O(n3) time. In contrast, our algorithm runs in time O~(ns2log⁡2n/Δmin⁡)\tilde{\mathcal{O}}(n s^2 \log^2 n / Δ_{\min})O~(ns2log2n/Δmin​), where Δmin⁡Δ_{\min}Δmin​ denotes the minimum spectral gap of the normalized matrix. For inverse gaps up to Δmin⁡−1=O(n)Δ_{\min}^{-1} = \mathcal{O}(n)Δmin−1​=O(n), the most favorable scaling for nnn distinct eigenvalues, attained by nearly uniformly spaced spectra such as those of perfect-state-transfer spin chains and the Sylvester-Kac matrix, and for sparsity up to s=o(n/log⁡n)s = o(\sqrt{n}/\log n)s=o(n​/logn), the proposed algorithm achieves an improved asymptotic runtime over the classical O(n3)\mathcal{O}(n^3)O(n3) approach for computing all eigenvalues and eigenvectors, under standard sparse-access assumptions.