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

TL;DR

This paper introduces two efficient quantum algorithms, S-FABLE and LS-FABLE, for approximate block-encoding of unstructured sparse matrices, significantly reducing quantum and classical resource requirements compared to previous methods.

Contribution

The paper presents novel modifications of the FABLE algorithm tailored for sparse matrices, achieving favorable scaling and resource efficiency in quantum circuit implementations.

Findings

S-FABLE requires approximately O(N) rotation gates for matrices with O(N) nonzero entries.

LS-FABLE reduces classical overhead by directly implementing sparse matrix entries.

Both methods demonstrate favorable scaling and resource reduction in quantum circuit complexity.

Abstract

The Fast Approximate BLock-Encoding algorithm (FABLE) is a technique to block-encode arbitrary $N\times N$ dense matrices into quantum circuits using at most $O(N^2)$ one and two-qubit gates and $\mathcal{O}(N^2\log{N})$ classical operations. The method nontrivially transforms a matrix $A$ into a collection of angles to be implemented in a sequence of $y$-rotation gates within the block-encoding circuit. If an angle falls below a threshold value, its corresponding rotation gate may be eliminated without significantly impacting the accuracy of the encoding. Ideally many of these rotation gates may be eliminated at little cost to the accuracy of the block-encoding such that quantum resources are minimized. In this paper we describe two modifications of FABLE to efficiently encode sparse matrices; in the first method termed Sparse-FABLE (S-FABLE), for a generic unstructured sparse matrix $A$ we use FABLE to block encode the Hadamard-conjugated matrix $H^{\otimes n}AH^{\otimes n}$ (computed with $\mathcal{O}(N^2\log N)$ classical operations) and conjugate the resulting circuit with $n$ extra Hadamard gates on each side to reclaim a block-approximation to $A$. We demonstrate that the FABLE circuits corresponding to block-encoding $H^{\otimes n}AH^{\otimes n}$ significantly compress and that overall scaling is empirically favorable (i.e. using S-FABLE to block-encode a sparse matrix with $\mathcal{O}(N)$ nonzero entries requires approximately $\mathcal{O}(N)$ rotation gates and $\mathcal{O}(N\log N)$ CNOT gates). In the second method called `Lazy' Sparse-FABLE (LS-FABLE), we eliminate the quadratic classical overhead altogether by directly implementing scaled entries of the sparse matrix $A$ in the rotation gates of the S-FABLE oracle. This leads to a slightly less accurate block-encoding than S-FABLE, while still demonstrating favorable scaling to FABLE similar to that found in S-FABLE.