Scholarlink

This study investigates the impact of depolarising noise on QAOA circuits, focusing on enhancing performance in solving the NP-Hard MaxCut problem. Using machine learning models to predict parameter sets, the Gaussian Process Regression model reduced the average and maximum iterations of the classical optimizer by 8.16% and 50%, and accelerated optimization by 13.02% on average and up to 37.65%. Training on a dataset with depolarising noise (0.01% on single qubit and 1% on two qubit gates) further reduced iterations by 28.33% and sped up optimization by 21.36%. The research offers new insights into noisy QAOA circuits and their application on real quantum hardware.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Noise-Aware Machine Learning for Accelerating QAOA in Noisy Intermediate-Scale Quantum Computing

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Poulami Paul,
  • Dipak Kumar Kole,
  • Debasmita Bhoumik,
  • Samik Mukherjee

摘要

This study investigates the impact of depolarising noise on QAOA circuits, focusing on enhancing performance in solving the NP-Hard MaxCut problem. Using machine learning models to predict parameter sets, the Gaussian Process Regression model reduced the average and maximum iterations of the classical optimizer by 8.16% and 50%, and accelerated optimization by 13.02% on average and up to 37.65%. Training on a dataset with depolarising noise (0.01% on single qubit and 1% on two qubit gates) further reduced iterations by 28.33% and sped up optimization by 21.36%. The research offers new insights into noisy QAOA circuits and their application on real quantum hardware.