Self assured Deep Learning with Minimum Pre Labeled Data for Wafer Pattern Classification

• Published in: IEEE transactions on semiconductor manufacturing Vol. 36; no. 3; p. 1
• Main Authors: Fan, Shu-Kai S., Tsai, Du-Ming, Shih, Ya-Fang
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Annotations; Artificial neural networks; Convolutional neural networks; Data models; Deep learning; Defects; Labeling; Labels; Machine learning; minimum pre-labeled data; Neural networks; Pattern classification; Pattern recognition; pseudo labeling; self-assured labeling; Semiconductor device modeling; Subject specialists; Supervised learning; Training; wafer defect pattern classification
• DOI: 10.1109/TSM.2023.3276816
• ISSN: 0894-6507

Data quality plays an important role during the training stage of machine/deep learning models. The annotation hinges on the experiences of domain experts. To acquire the experts knowledge in the context of machine learning, manual data labeling, a tedious and time-consuming task in supervised learning, should be given a top priority. However, the domain experts in the line of plentiful manual annotation may easily get distracted or fatigued after long-time work, causing judgment errors, mislabeling, etc. The pattern recognition of wafer defect map is investigated in this paper, the primary goal of which is to train the convolutional neural network (CNN) model through a very limited number of manually labeled data so that the trained model is capable of performing pseudo labeling. Subsequently, a self-assured adaptive ensemble learner in terms of a series of shallow neural networks is proposed to filter wafer map samples with untrusted pseudo-labels. In the result, the amount of human annotations is significantly reduced by 61% for training a highly accurate classifier. A minimum number of manually labeled data is suggested while the equally high classification performance of wafer defect pattern is maintained. For the evaluation purpose, the proposed self-assured learning is compared with the confidence learning.