1. [Publications](/publications)
2. Physics-Informed Optical Kernel Regression Using Complex-valued Neural Fields
 
 # Physics-Informed Optical Kernel Regression Using Complex-valued Neural Fields

  ![](/sites/default/files/styles/wide/public/publications/Weixin%20Screenshot_20230606115217.png?itok=XqagZf6-)

 Lithography is fundamental to integrated circuit fabrication, necessitating large computation overhead. The advancement of machine learning (ML)-based lithography models alleviates the trade-offs between manufacturing process expense and capability. However, all previous methods regard the lithography system as an image-to-image black box mapping, utilizing network parameters to learn by rote mappings from massive mask-to-aerial or mask-to-resist image pairs, resulting in poor generalization capability. In this paper, we propose a new ML-based paradigm disassembling the rigorous lithographic model into non-parametric mask operations and learned optical kernels containing determinant source, pupil, and lithography information. By optimizing complex-valued neural fields to perform optical kernel regression from coordinates, our method can accurately restore lithography system using a small-scale training dataset with fewer parameters, demonstrating superior generalization capability as well. Experiments show that our framework can use 31% of parameters while achieving 69× smaller mean squared error with 1.3× higher throughput than the state-of-the-art.



 ## Authors



Guojin Chen (CUHK)

Zehua Pei (CUHK)

[Haoyu Yang](/person/haoyu-yang)

Yuzhe Ma (HKUST)

Bei Yu (CUHK)

Martin Wong (CUHK)

 

 

 ## Publication Date



Thursday, July 6, 2023

 

 ## Published in



[60th ACM/IEEE Design Automation Conference](https://arxiv.org/abs/2303.08435)

 

 ## Research Area



[Artificial Intelligence and Machine Learning ](/research-area/machine-learning-artificial-intelligence)

[Circuits and VLSI Design](/research-area/circuits)

[Generative AI](/research-area/generative-ai)

 

 

 ## Uploaded Files



[2303.08435.pdf](https://d1qx31qr3h6wln.cloudfront.net/publications/2303.08435.pdf "Open file in new window")2.66 MB