Abstract:We propose fast and practical quantum-inspired classical algorithms for solving linear systems. Specifically, given sampling and query access to a matrix $A\in\mathbb{R}^{m\times n}$ and a vector $b\in\mathbb{R}^m$, we propose classical algorithms that produce a data structure for the solution $x\in\mathbb{R}^{n}$ of the linear system $Ax=b$ with the ability to sample and query its entries. The resulting $x$ satisfies $\|x-A^{+}b\|\leq\epsilon\|A^{+}b\|$, where $\|\cdot\|$ is the spectral norm and $A^+$ is the Moore-Penrose inverse of $A$. Our algorithm has time complexity $\widetilde{O}(\kappa_F^4/\kappa\epsilon^2)$ in the general case, where $\kappa_{F} =\|A\|_F\|A^+\|$ and $\kappa=\|A\|\|A^+\|$ 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 $\widetilde{O}(s \kappa\log(1/\epsilon))$, matching the quantum lower bound for solving linear systems in $\kappa$ and $1/\epsilon$ up to poly-logarithmic factors [Harrow and Kothari]. When $A$ is $s$-sparse and symmetric positive-definite, our algorithm has complexity $\widetilde{O}(s\sqrt{\kappa}\log(1/\epsilon))$. 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.

A square-root speedup for finding the smallest eigenvalue

Quantum Algorithm For Estimating Eigenvalue

A Quantum Speed-Up for Approximating the Top Eigenvectors of a Matrix

Quantum speedups for linear programming via interior point methods

A Query-based Quantum Eigensolver

Quantum eigenvalue processing

Algorithmic error mitigation for quantum eigenvalues estimation

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

Quantum Algorithm for a Convergent Series of Approximations towards the Exact Solution of the Lowest Eigenstates of a Hamiltonian

Fast and Practical Quantum-Inspired Classical Algorithms for Solving Linear Systems.

Eigenpath traversal by Poisson-distributed phase randomisation

Quantum linear system algorithm with optimal queries to initial state preparation

Solving eigenvalue problems obtained by the finite element method on a quantum annealer using only a few qubits

Efficient State Read-out for Quantum Machine Learning Algorithms

High-precision and low-depth eigenstate property estimation: theory and resource estimation

Quantum Data Fitting Algorithm for Non-sparse Matrices

A shortcut to an optimal quantum linear system solver

A Quadratic Speedup in Finding Nash Equilibria of Quantum Zero-Sum Games

A Novel Quantum-Classical Hybrid Algorithm for Determining Eigenstate Energies in Quantum Systems

Real-Time Krylov Theory for Quantum Computing Algorithms