Solving Differential Equations via Continuous-Variable Quantum Computers

Abstract

We explore how a continuous-variable (CV) quantum computer could solve a classic differential equation, making use of its innate capability to represent real numbers in qumodes. Specifically, we construct variational CV quantum circuits [Killoran et al., Phys.~Rev.~Research 1, 033063 (2019)] to approximate the solution of one-dimensional ordinary differential equations (ODEs), with input encoding based on displacement gates and output via measurement averages. Our simulations and parameter optimization using the PennyLane / Strawberry Fields framework demonstrate good convergence for both linear and non-linear ODEs.

arXiv comment6 pages, 8 figures

References in corpus (9)

• Quantum information with continuous variables
• Gaussian Quantum Information
• Quantum Circuit Learning
• On the Convergence of Adam and Beyond
• Quantum computation over continuous variables
• Evaluating analytic gradients on quantum hardware
• Strawberry Fields: A Software Platform for Photonic Quantum Computing
• Efficient quantum algorithm for dissipative nonlinear differential equations
• PennyLane: Automatic differentiation of hybrid quantum-classical computations