Topic

Published in Quantum

Research on "Efficient Expectation Evaluation by Quantum Computers" jointly conducted by TOYOTA CENTRAL R&D LABS., INC. and TOYOTA MOTOR CORPORATION was published in Quantum.

In a quantum computer, information is obtained stochastically by measuring qubits that have been operated several times, and various quantities are calculated using the information. Because both the number of qubits operations and measurements are directly related to the computational complexity, both of them required to be small. In this paper, we propose a new method that significantly reduces the number of measurements required to evaluate the expected value, which is one of the most fundamental quantum subroutine, with only a few additional quantum operations. Experimental verification using the IBM-Q showed that the expected value can be evaluated with exponentially fewer measurements than existing methods under specific conditions. Expectation evaluation is used in many quantum algorithms such as quantum chemical calculations, optimization, and quantum machine learning. The proposed method contributes to speeding up these calculations.
Our code is available at
https://github.com/ToyotaCRDL/extended-bell-measurements

Title: Computationally Efficient Quantum Expectation with Extended Bell Measurements
Authors:Kondo, R., Sato, Y., Koide, S., Kajita, S., Takamatsu, H.
Journal Name: Quantum
Published: 13 April 2022
https://doi.org/10.22331/q-2022-04-13-688

Back to list
PAGE TOP