GEM-QAS: A Gate-Efficient Evolutionary Framework for Scalable Quantum Architecture Search in Intelligent Industrial Informatics
Chuang Wu, Zhulong Diao, Hongding Zhang, Cangming Liang, Naixue N. Xiong, Wei Liang
Hunan University of Science and Technology University of South China
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
Evolutionary quantum architecture search (QAS) automates efficient ansatz design for intelligent and industrial applications. However, current methods suffer from inadequate candidate exploration due to limited evaluation epochs and experience late-stage evolutionary instability caused by disruptive random mutations. To address these challenges, this article proposes gate-efficient mutation-driven QAS. Specifically, an adaptive binary tournament selection algorithm is designed to dynamically balance exploration and exploitation. This mechanism enhances convergence efficiency by prioritizing structurally promising ansätze in the early stages while enforcing fitness-driven selection in later iterations. Furthermore, an inheritable mutation algorithm is designed to preserve beneficial circuit structures from parent architectures while introducing adaptive mutations, thereby sustaining population diversity and ensuring search stability. Experiments across multiple standard datasets indicate that this method improves classification accuracy by 5%–18% over state-of-the-art baselines. Finally, the method achieves highly effective defect detection on the NEU-DET dataset using minimal quantum gates. This gate reduction successfully mitigates quantum noise and highlights the potential of the method for robust industrial deployment.
逐年被引趋势
暂无年度引用数据
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
计算机 / AIQuantum Computing Algorithms and Architecture
Evolutionary Algorithms and Applications · Metaheuristic Optimization Algorithms Research
参考文献 35
此处列出前 3 条