arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2111.06982cs.LGcs.AI
收藏

软测量模型可视化:从模型学习中微调神经网络

Soft Sensing Model Visualization: Fine-tuning Neural Network from What Model Learned

Seagate Technology(希捷科技);Case Western Reserve University(凯斯西储大学);University of Chicago(芝加哥大学);University of Cincinnati(辛辛那提大学);Florida Atlantic University(佛罗里达大西洋大学);Duke Kunshan University(昆山杜克大学)

Xiaoye Qian, Chao Zhang, Jaswanth Yella, Yu Huang, Ming-Chun Huang, Sthitie Bom

摘要:

从智能制造中收集的数据的可用性越来越高,正在改变生产监控和控制的模式。除了随时间变化的意外干扰和不确定性之外,晶片制造过程的复杂性和内容的增加使得利用基于模型的方法进行控制过程是不可行的。因此,数据驱动的软测量建模在晶圆工艺诊断中变得更加普遍。近年来,深度学习在处理高度非线性和动态的时间序列数据时表现出良好的性能,已被应用于软测量系统中。然而,尽管在软测量系统中取得了成功,但深度学习框架的底层逻辑很难理解。在本文中,我们提出了一个基于深度学习的模型,用于使用高度不平衡的数据集进行缺陷晶圆检测。为了理解所提出的模型是如何工作的,应用了深度可视化方法。此外,模型然后在深度可视化的指导下进行微调。大量的实验进行验证所提出的系统的有效性。结果提供了一个解释模型的工作原理和指导性的微调方法的基础上的解释。

英文摘要:

The growing availability of the data collected from smart manufacturing is changing the paradigms of production monitoring and control. The increasing complexity and content of the wafer manufacturing process in addition to the time-varying unexpected disturbances and uncertainties, make it infeasible to do the control process with model-based approaches. As a result, data-driven soft-sensing modeling has become more prevalent in wafer process diagnostics. Recently, deep learning has been utilized in soft sensing system with promising performance on highly nonlinear and dynamic time-series data. Despite its successes in soft-sensing systems, however, the underlying logic of the deep learning framework is hard to understand. In this paper, we propose a deep learning-based model for defective wafer detection using a highly imbalanced dataset. To understand how the proposed model works, the deep visualization approach is applied. Additionally, the model is then fine-tuned guided by the deep visualization. Extensive experiments are performed to validate the effectiveness of the proposed system. The results provide an interpretation of how the model works and an instructive fine-tuning method based on the interpretation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑