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.

Explicit Quantum Circuits for Block Encodings of Certain Sparse Matrices

Block encoding of sparse structured matrices coming from ocean acoustics in quantum computing

Block-encoding structured matrices for data input in quantum computing

The power of block-encoded matrix powers: improved regression techniques via faster Hamiltonian simulation

FABLE: Fast Approximate Quantum Circuits for Block-Encodings

Circuit complexity of quantum access models for encoding classical data

Quantum algorithms for matrix operations and linear systems of equations

Block encoding of matrix product operators

Efficient quantum circuits for dense and non-unitary operators

S-FABLE and LS-FABLE: Fast approximate block-encoding algorithms for unstructured sparse matrices

Block encoding by signal processing

Nonlinear quantum computing by amplified encodings

Quantum sampling algorithms for quantum state preparation and matrix block-encoding

Quantum Signal Processing and Quantum Singular Value Transformation on $U(N)$

Double-Logarithmic Depth Block-Encodings of Simple Finite Difference Method's Matrices

Explicit block encodings of boundary value problems for many-body elliptic operators

Improved Quantum Algorithms for Eigenvalues Finding and Gradient Descent

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