隨著半導體快速的發展，熱點偵測被視為非常重要且有挑戰性的任務。在最近幾年，使用深度學習的方法成功地得到了不錯的效果。然而，大部分的作法都太專注於給定的訓練資料，導致偵測的效果單一，只能應用於競賽之中。本篇提出了新的熱點偵測方法，不同於以往的訓練模式，我們使用光刻模型來得到曝光過後的形變圖，藉此找出熱點發生的成因，得到一個泛用性更高的模型，在實際的應用上更加精準。我們也針對了目前的曝光模擬技術加以改良，提出更可靠的模型，使得整體預測結果更加完美且穩定。透過產生的形變資訊，我們的模型有能力去找出發生熱點的特徵與正常電路的不同，就可以偵測出更多潛在的瑕疵設計，而不是只會找出原先給定的熱點。後續一連串的實驗結果也證明了我們架構的可行性與優點，比較了跟以往熱點偵測模型的不同。

With the rapid advancement of semiconductors, hotspot detection is regarded as a challenging and crucial task in the design flow. Deep learning-based methods have been implemented in recent years, and great success has been achieved. Nonetheless, most works can only detect the corresponding hotspots in the given training pairs. In this paper, we first propose a novel detection method where a lithography simulator and hotspot detection are implemented simultaneously. First of all, we improve the current lithography simulator model to make the overall prediction results perfect and stable. Based on the deformation map from our model, it is capable of finding what causes the hotspots and detecting potential problematic weak patterns. We focus on variations in energy levels instead of given hotspots like in previous works. A series of experiments demonstrate that our framework carries out more advantages indeed.

摘要 i
Abstract ii
1 Introduction 1
2 Related Work 4
2.1 Lithography simulation models 4
2.2 Hotspot detection 6
2.3 Attention modules 8
3 Proposed Method 11
3.1 Overview 11
3.2 Enhanced Lithography Simulator 12
3.3 Detection Network Architecture 14
a Deformation map 15
b Derivative Module 17
c Deformation Fusion Module 19
d Loss function 20
e Risk score of Hotspot 21
4 Experiment 23
4.1 Datasets and Network Configuration 23
4.2 Performance Metrics 25
4.3 Experimental results 27
a Enhanced-LithoNet 27
b Evaluation of Risk Score 29
c The performance on Dataset-2 29
d The performance on Dataset-3 30
e Ablation study on detection network 32
f Ablation study on loss function 33
5 Conclusion 34
References 35

[1] H.-C. Shao, C.-Y. Peng, J.-R. Wu, C.-W. Lin, S.-Y. Fang, P.-Y. Tsai, and
Y.-H. Liu, “From ic layout to die photograph: A cnn-based data-driven
approach,” IEEE Transactions on Computer-Aided Design of Integrated
Circuits and Systems, vol. 40, no. 5, pp. 957–970, 2020.
[2] Y. Watanabe, T. Kimura, T. Matsunawa, and S. Nojima, “Accurate lithog-
raphy simulation model based on convolutional neural networks,” in Opti-
cal Microlithography XXX, vol. 10147, pp. 137–145, SPIE, 2017.
[3] W. Ye, M. B. Alawieh, Y. Lin, and D. Z. Pan, “Lithogan: End-to-end
lithography modeling with generative adversarial networks,” in 2019 56th
ACM/IEEE Design Automation Conference (DAC), pp. 1–6, IEEE, 2019.
[4] S.-Y. Lin, J.-Y. Chen, J.-C. Li, W.-Y. Wen, and S.-C. Chang, “A novel
fuzzy matching model for lithography hotspot detection,” in 2013 50th
ACM/EDAC/IEEE Design Automation Conference (DAC), pp. 1–6, IEEE,
2013.
[5] W.-Y. Wen, J.-C. Li, S.-Y. Lin, J.-Y. Chen, and S.-C. Chang, “A fuzzy-
matching model with grid reduction for lithography hotspot detection,”
IEEE Transactions on Computer-Aided Design of Integrated Circuits and
Systems, vol. 33, no. 11, pp. 1671–1680, 2014.
[6] Y.-T. Yu, Y.-C. Chan, S. Sinha, I. H.-R. Jiang, and C. Chiang, “Accurate
process-hotspot detection using critical design rule extraction,” in Proceed-
ings of the 49th Annual Design Automation Conference, pp. 1167–1172,
2012.
[7] B. Zhu, R. Chen, X. Zhang, F. Yang, X. Zeng, B. Yu, and M. D. Wong,
“Hotspot detection via multi-task learning and transformer encoder,” in
2021 IEEE/ACM International Conference On Computer Aided Design
(ICCAD), pp. 1–8, 2021.
[8] R. Chen, W. Zhong, H. Yang, H. Geng, X. Zeng, and B. Yu, “Faster region-
based hotspot detection,” in 2019 56th ACM/IEEE Design Automation
Conference (DAC), pp. 1–6, 2019.
[9] R. Girshick, “Fast r-cnn,” in Proceedings of the IEEE international con-
ference on computer vision, pp. 1440–1448, 2015
[10] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez,
Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neu-
ral information processing systems, vol. 30, 2017.
[11] H. Zhang, I. Goodfellow, D. Metaxas, and A. Odena, “Self-attention gener-
ative adversarial networks,” in International conference on machine learn-
ing, pp. 7354–7363, PMLR, 2019.
[12] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image
recognition,” in Proceedings of the IEEE conference on computer vision
and pattern recognition, pp. 770–778, 2016.
[13] Y. Jiang, F. Yang, B. Yu, D. Zhou, and X. Zeng, “Efficient layout hotspot
detection via binarized residual neural network ensemble,” IEEE Trans-
actions on Computer-Aided Design of Integrated Circuits and Systems,
vol. 40, no. 7, pp. 1476–1488, 2021.
[14] S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “Cbam: Convolutional block
attention module,” in Proceedings of the European conference on computer
vision (ECCV), pp. 3–19, 2018.
[15] O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks
for biomedical image segmentation,” in Medical Image Computing and
Computer-Assisted Intervention–MICCAI 2015: 18th International Con-
ference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18,
pp. 234–241, Springer, 2015.
[16] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality
assessment: from error visibility to structural similarity,” IEEE transac-
tions on image processing, vol. 13, no. 4, pp. 600–612, 2004.
[17] T. Gai, T. Qu, S. Wang, X. Su, R. Xu, Y. Wang, J. Xue, Y. Su, Y. Wei, and
T. Ye, “Flexible hotspot detection based on fully convolutional network
with transfer learning,” IEEE Transactions on Computer-Aided Design of
Integrated Circuits and Systems, 2021.