QAOA’s Circuit Optimization And Construction Of Distributed Simulation Platform

The QAOA algorithm is one of the most popular quantum algorithms in recent years,which has received widespread attention and research since its proposal.This algorithm aims to solve combinatorial optimization problems,and compared with classical optimization algorithms,QAOA has a significant acceleration effect and can solve a wider range of combinatorial optimization problems.When encountering large-scale problems,QAOA can handle problems that classical algorithms are difficult to solve and better demonstrate quantum advantages.The algorithm has three aspects that can be optimized:1.The core of the QAOA algorithm is the quantum circuit.In the initial QAOA algorithm,the number of CNOT gates in the Ansatz circuit is determined by the number of edges in the combinatorial optimization problem.However,in the mapping of vertex-edge operations,useless CNOT gates can be removed to reduce noise errors in real quantum environments.2.The performance of the QAOA algorithm also depends on the classical optimizer,which runs on classical computers.In the classical computer algorithm part,specific computer hardware devices can be used to optimize and improve the accuracy of the algorithm’s results.3.Because today’s NISQ resources are limited,most of the applications of this algorithm are still in the framework simulation stage.There is still a lot of room for improvement in the efficiency and optimization accuracy of the framework simulation.Regarding these possibilities for optimization,the research carried out in this paper is as follows:1.Propose a QAOA circuit optimization algorithm for multiple graph scenarios.Based on the specific graph scenario of the QAOA maximum cut problem and the circuit construction for solving the maximum cut problem,we propose a QAOA circuit optimization algorithm for multiple graph scenarios to eliminate removable CNOT gates.We use the Qiskit quantum framework to simulate the QAOA maximum cut problem before and after optimization for multiple graph scenarios and verify the advantage of the circuit optimization algorithm under noisy real quantum device environment using the IBM Quantum Composer.Experimental results show that the optimized algorithm can reduce the number of CNOT gates by nearly 40%in most cases,and the operation of the algorithm is more stable under real noise environment.2.Propose a parallel computation optimization algorithm based on Qiskit’s QAOA.The speed of the path optimization algorithm is greatly affected when dealing with a large number of qubits.Therefore,the specific practical graph is split into independent subgraphs that can run independently,and the circuits are optimized in parallel.In addition,parallel design is applied to the SPSA optimizer and applied to problem solving.Finally,experimental comparisons were conducted,and the results showed that the optimization improved significantly with the increase of qubits under 3 logical CPUs,and the speed of circuit optimization for 1500 qubits increased by 2.2 times.The optimized SPSA improved the average operation results by 50%at 10 and 20 qubits,closer to the ideal expected value.These studies aim to improve the simulation efficiency of the algorithm,shorten the computation time,and prepare for the future application of classical optimizers to quantum circuits.3.Independently design and develop a distributed simulation platform based on Qiskit.In order to extend the research of quantum algorithms,we have developed a distributed simulation platform based on Qiskit.The platform uses Rabbit MQ for distributed information acquisition and Vue and Flask for front-end and back-end technologies to implement a web platform.In the independently developed simulation platform,practical problems can be solved through the application of the QAOA algorithm,and the selection of problem size and code editing can be realized,and the quantum circuit diagram,classical computation perturbation process diagram,and result display diagram can be displayed in real-time.In addition,this platform combines the most advanced container orchestration technology Kubernetes to achieve cloud-native application deployment design,and can seamlessly integrate with public cloud service providers such as AWS,Azure,and Alibaba Cloud.