A Self-Supervised Learning of a Foundation Model for Analog Layout Design Automation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Jeong, Sungyu, Choi, Won Joon, Choi, Junung, Biswas, Anik, Kim, Byungsub
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918216422391808
author Jeong, Sungyu
Choi, Won Joon
Choi, Junung
Biswas, Anik
Kim, Byungsub
author_facet Jeong, Sungyu
Choi, Won Joon
Choi, Junung
Biswas, Anik
Kim, Byungsub
contents We propose a UNet-based foundation model and its self-supervised learning method to address two key challenges: 1) lack of qualified annotated analog layout data, and 2) excessive variety in analog layout design tasks. For self-supervised learning, we propose random patch sampling and random masking techniques automatically to obtain enough training data from a small unannotated layout dataset. The obtained data are greatly augmented, less biased, equally sized, and contain enough information for excessive varieties of qualified layout patterns. By pre-training with the obtained data, the proposed foundation model can learn implicit general knowledge on layout patterns so that it can be fine-tuned for various downstream layout tasks with small task-specific datasets. Fine-tuning provides an efficient and consolidated methodology for diverse downstream tasks, reducing the enormous human effort to develop a model per task separately. In experiments, the foundation model was pre-trained using 324,000 samples obtained from 6 silicon-proved manually designed analog circuits, then it was fine-tuned for the five example downstream tasks: generating contacts, vias, dummy fingers, N-wells, and metal routings. The fine-tuned models successfully performed these tasks for more than one thousand unseen layout inputs, generating DRC/LVS-clean layouts for 96.6% of samples. Compared with training the model from scratch for the metal routing task, fine-tuning required only 1/8 of the data to achieve the same dice score of 0.95. With the same data, fine-tuning achieved a 90% lower validation loss and a 40% higher benchmark score than training from scratch.
format Preprint
id arxiv_https___arxiv_org_abs_2503_22143
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Self-Supervised Learning of a Foundation Model for Analog Layout Design Automation
Jeong, Sungyu
Choi, Won Joon
Choi, Junung
Biswas, Anik
Kim, Byungsub
Signal Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
We propose a UNet-based foundation model and its self-supervised learning method to address two key challenges: 1) lack of qualified annotated analog layout data, and 2) excessive variety in analog layout design tasks. For self-supervised learning, we propose random patch sampling and random masking techniques automatically to obtain enough training data from a small unannotated layout dataset. The obtained data are greatly augmented, less biased, equally sized, and contain enough information for excessive varieties of qualified layout patterns. By pre-training with the obtained data, the proposed foundation model can learn implicit general knowledge on layout patterns so that it can be fine-tuned for various downstream layout tasks with small task-specific datasets. Fine-tuning provides an efficient and consolidated methodology for diverse downstream tasks, reducing the enormous human effort to develop a model per task separately. In experiments, the foundation model was pre-trained using 324,000 samples obtained from 6 silicon-proved manually designed analog circuits, then it was fine-tuned for the five example downstream tasks: generating contacts, vias, dummy fingers, N-wells, and metal routings. The fine-tuned models successfully performed these tasks for more than one thousand unseen layout inputs, generating DRC/LVS-clean layouts for 96.6% of samples. Compared with training the model from scratch for the metal routing task, fine-tuning required only 1/8 of the data to achieve the same dice score of 0.95. With the same data, fine-tuning achieved a 90% lower validation loss and a 40% higher benchmark score than training from scratch.
title A Self-Supervised Learning of a Foundation Model for Analog Layout Design Automation
topic Signal Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2503.22143