Physics-Informed Optical Kernel Regression Using Complex-valued Neural Fields
This addresses the trade-offs between manufacturing process expense and capability in lithography, offering improved generalization and efficiency for integrated circuit fabrication.
The paper tackles the problem of high computational overhead in lithography for integrated circuit fabrication by proposing a new machine learning-based paradigm that disassembles the lithographic model into non-parametric mask operations and learned optical kernels, achieving 69× smaller mean squared error with 31% of parameters and 1.3× higher throughput compared to state-of-the-art methods.
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$\times$ smaller mean squared error with 1.3$\times$ higher throughput than the state-of-the-art.