Presentation
Physics-Informed Optical Kernel Regression Using Complex-valued Neural Fields
DescriptionLithography is the fundamental part of integrated circuit fabrication, which typically accounts for around 30% of the cost of manufacturing a chip. The advancement of deep learning-based lithography models alleviates the trade-offs between manufacturing process expense and capability. However, all previous methods regard the lithography system as an end-to-end black box. In this paper, we investigate the underlying mathematical representation of lithographic imaging model and design a physics-informed neural network for lithography simulation. By optimizing complex-valued neural fields to predict transmission cross-coefficient (TCC) kernels, our method can accurately restore lithography system using a small-scale training dataset with the fewest parameters.
Event Type
Research Manuscript
TimeWednesday, July 12th1:40pm - 1:55pm PDT
Location3008, 3rd Floor
EDA
Design for Manufacturing and Reliability