Self-Attention to Operator Learning-based 3D-IC Thermal Simulation

Thermal management in 3D ICs is increasingly challenging due to higher power densities. Traditional PDESolving based methods, while accurate, are too slow for iterative design. Machine learning approaches like FNO provide faster alternatives but suffer from high-frequency information loss and high-f...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	2025 62nd ACM/IEEE Design Automation Conference (DAC) s. 1 - 7
Hlavní autoři:	Huang, Zhen, Wang, Hong, Yang, Wenkai, Tang, Muxi, Xie, Depeng, Lin, Ting-Jung, Zhang, Yu, Xing, Wei W., He, Lei
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	IEEE 22.06.2025
Témata:	Accuracy Computational modeling Fourier Neural Operator Integrated circuit modeling Self-Attention Solid modeling Thermal analysis Thermal expansion Thermal management Thermal modeling Three-dimensional displays Training Transfer learning
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Thermal management in 3D ICs is increasingly challenging due to higher power densities. Traditional PDESolving based methods, while accurate, are too slow for iterative design. Machine learning approaches like FNO provide faster alternatives but suffer from high-frequency information loss and high-fidelity data dependency. We introduce Self-Attention UNet Fourier Neural Operator (SAU-FNO), a novel framework combining self-attention and U-Net with FNO to capture longrange dependencies and model local high-frequency features effectively. Transfer learning is employed to fine-tune low-fidelity data, minimizing the need for extensive high-fidelity datasets and speeding up training. Experiments demonstrate that SAUFNO achieves state-of-the-art thermal prediction accuracy and provides an 842 \times speedup over traditional FEM methods, making it an efficient tool for advanced 3D IC thermal simulations.
DOI:	10.1109/DAC63849.2025.11132988