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.

Quantum Multigrid Algorithm for Finite Element Problems

Quantum Relaxation for Linear Systems in Finite Element Analysis

qRLS: quantum relaxation for linear systems in finite element analysis

Hierarchical Multigrid Ansatz for Variational Quantum Algorithms

Quantum multi-row iteration algorithm for linear systems with non-square coefficient matrices

An Efficient Quantum Algorithm for Linear System Problem in Tensor Format

Quantum conjugate gradient method using the positive-side quantum eigenvalue transformation

Improvements to Quantum Interior Point Method for Linear Optimization

Efficient State Read-out for Quantum Machine Learning Algorithms

Quantum Multi-Resolution Measurement with application to Quantum Linear Solver

Quantum Algorithm for Solving a Quadratic Nonlinear System of Equations

Efficient quantum linear solver algorithm with detailed running costs

Quantum Gram-Schmidt Processes and Their Application to Efficient State Readout for Quantum Algorithms

Towards Quantum Computational Mechanics

Quantum Krylov-Subspace Method Based Linear Solver

Quantum Realization of the Finite Element Method

Quantum Newton's method for solving system of nonlinear algebraic equations

Quantum Discrete Adiabatic Linear Solver based on Block Encoding and Eigenvalue Separator

Quantum Algorithms for Ground-State Preparation and Green's Function Calculation

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

An Implementation of the Finite Element Method in Hybrid Classical/Quantum Computers