Enhanced Active Learning of Convolutional Neural Networks: A Case Study for Defect Classification in the Semiconductor Industry

Georgios Koutroulis, Tiago Santos, Michael Wiedemann, Christian Faistauer, Roman Kern, Roman Kern, Stefan Thalmann

2020

Abstract

With the advent of high performance computing and scientific advancement, deep convolutional neural networks (CNN) have already been established as the best candidate for image classification tasks. A decisive requirement for successful deployment of CNN models is the vast amount of annotated images, which usually is a costly and quite tedious task, especially within an industrial environment. To address this deployment barrier, we propose an enhanced active learning framework of a CNN model with a compressed architecture for chip defect classification in semiconductor wafers. Our framework unfolds in two main steps and is performed in an iterative manner. First, a subset of the most informative samples is queried based on uncertainty estimation. Second, spatial metadata of the queried images are utilized for a density-based clustering in order to discard noisy instances and to keep only those ones that constitute systematic defect patterns in the wafer. Finally, a reduced and more representative subset of images are passed for labelling, thus minimizing the manual labour of the process engineer. In each iteration, the performance of the CNN model is considerably improved, as only those images are labeled that will help the model to better generalize. We validate the effectiveness of our framework using real data from running processes of a semiconductor manufacturer.

Paper Citation

in Harvard Style

Koutroulis G., Santos T., Wiedemann M., Faistauer C., Kern R. and Thalmann S. (2020). Enhanced Active Learning of Convolutional Neural Networks: A Case Study for Defect Classification in the Semiconductor Industry. In Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - Volume 1: KDIR; ISBN 978-989-758-474-9, SciTePress, pages 269-276. DOI: 10.5220/0010142902690276

in Bibtex Style

@conference{kdir20,
author={Georgios Koutroulis and Tiago Santos and Michael Wiedemann and Christian Faistauer and Roman Kern and Stefan Thalmann},
title={Enhanced Active Learning of Convolutional Neural Networks: A Case Study for Defect Classification in the Semiconductor Industry},
booktitle={Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - Volume 1: KDIR},
year={2020},
pages={269-276},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010142902690276},
isbn={978-989-758-474-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - Volume 1: KDIR
TI - Enhanced Active Learning of Convolutional Neural Networks: A Case Study for Defect Classification in the Semiconductor Industry
SN - 978-989-758-474-9
AU - Koutroulis G.
AU - Santos T.
AU - Wiedemann M.
AU - Faistauer C.
AU - Kern R.
AU - Thalmann S.
PY - 2020
SP - 269
EP - 276
DO - 10.5220/0010142902690276
PB - SciTePress