L-QLES: Sparse Laplacian generator for evaluating Quantum Linear Equation Solvers
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916131513565184 |
|---|---|
| author | Lapworth, Leigh |
| author_facet | Lapworth, Leigh |
| contents | L-QLES is an open source python code for generating 1D, 2D and 3D Laplacian operators and associated Poisson equations and their classical solutions. Its goal is to provide quantum algorithm developers with a flexible test case framework where features of industrial applications can be incorporated without the need for end-user domain knowledge or reliance on inflexible one-off industry supplied matrix sets. A sample set of 1, 2, and 3 dimensional Laplacians are suggested and used to compare the performance of the Prepare-Select and FABLE block encoding techniques. Results show that large matrices are not needed to investigate industrial characteristics. A matrix with a condition number of 17,000 can be encoded using 13 qubits. L-QLES has also been produced to enable algorithm developers to investigate and optimise both the classical and quantum aspects of the inevitable hybrid nature of quantum linear equation solvers. Prepare-Select encoding that takes over an hour of classical preprocessing time to decompose a 4,096x4,096 matrix into Pauli strings can be can investigated using L-QLES matrices. Similarly, row-column query oracles that have success probabilities $\le 10^{-7}$ for the same matrix can be investigated. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_12266 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | L-QLES: Sparse Laplacian generator for evaluating Quantum Linear Equation Solvers Lapworth, Leigh Quantum Physics L-QLES is an open source python code for generating 1D, 2D and 3D Laplacian operators and associated Poisson equations and their classical solutions. Its goal is to provide quantum algorithm developers with a flexible test case framework where features of industrial applications can be incorporated without the need for end-user domain knowledge or reliance on inflexible one-off industry supplied matrix sets. A sample set of 1, 2, and 3 dimensional Laplacians are suggested and used to compare the performance of the Prepare-Select and FABLE block encoding techniques. Results show that large matrices are not needed to investigate industrial characteristics. A matrix with a condition number of 17,000 can be encoded using 13 qubits. L-QLES has also been produced to enable algorithm developers to investigate and optimise both the classical and quantum aspects of the inevitable hybrid nature of quantum linear equation solvers. Prepare-Select encoding that takes over an hour of classical preprocessing time to decompose a 4,096x4,096 matrix into Pauli strings can be can investigated using L-QLES matrices. Similarly, row-column query oracles that have success probabilities $\le 10^{-7}$ for the same matrix can be investigated. |
| title | L-QLES: Sparse Laplacian generator for evaluating Quantum Linear Equation Solvers |
| topic | Quantum Physics |
| url | https://arxiv.org/abs/2402.12266 |