Work

QAOAKit: A Toolkit for Reproducible Study, Application, and Verification of the QAOA

Year: 2021

Type: conference-paper

Abstract: Understanding the best known parameters, performance, and systematic behavior of the Quantum Approximate Optimization Algorithm (QAOA) remain open research questions, even as the algorithm gains popularity. We introduce QAOAKit, a Python toolkit for the QAOA built for exploratory research. QAOAKit is a unified repository of preoptimized QAOA parameters and circuit generators for common quantum simulation frameworks. We combine, standardize, and cross-validate previously known parameters for the MaxCut problem, and incorporate this into QAOAKit. We also build conversion tools to use these parameters as inputs in several quantum simulation frameworks that can be used to reproduce, compare, and extend known results from various sources in the literature. We describe QAOAKit and provide examples of how it can be used to reproduce research results and tackle open problems in quantum optimization.

Source: IEEE/ACM International Workshop on Quantum Computing Software (QCS)

DOI: 10.1109/qcs54837.2021.00011

Authors: Ruslan Shaydulin, Kunal Marwaha, Jonathan Wurtz, Phillip C. Lotshaw

Institutions: Argonne National Laboratory, University of Chicago, University of California, Berkeley, Tufts University, Oak Ridge National Laboratory

Language: English

Topic: Quantum Computing Algorithms and Architecture

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green