Deep Hierarchical Knowledge Loss for Fault Intensity Diagnosis
深度层次知识损失用于故障强度诊断
机构 * School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)科学与工程学院) ; School of Artificial Intelligence, The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)人工智能学院) ; School of Artificial Intelligence, Xidian University(西安电子科技大学人工智能学院) ; School of Computer, Luoyang Institute of Science and Technology(洛阳理工学院计算机学院) ; Frankfurt Institute for Advanced Studies(法兰克福先进研究 institute) ; SAMSON AG(SAMSON公司)
专题命中 诊断辅助 :diagnosis(title,abstract);分类 cs.CV、cs.LG、eess.SP
AI总结 本文提出深度层次知识损失框架,通过层次树损失和焦点层次树损失提升故障诊断的识别能力,实验表明在多个工业数据集上优于现有方法。
Comments The paper has been accepted by Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.1 (KDD 2026)