Learning Library Cell Representations in Vector Space

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liang, Rongjian, Lu, Yi-Chen, Liu, Wen-Hao, Ren, Haoxing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908289770455040
author Liang, Rongjian
Lu, Yi-Chen
Liu, Wen-Hao
Ren, Haoxing
author_facet Liang, Rongjian
Lu, Yi-Chen
Liu, Wen-Hao
Ren, Haoxing
contents We propose Lib2Vec, a novel self-supervised framework to efficiently learn meaningful vector representations of library cells, enabling ML models to capture essential cell semantics. The framework comprises three key components: (1) an automated method for generating regularity tests to quantitatively evaluate how well cell representations reflect inter-cell relationships; (2) a self-supervised learning scheme that systematically extracts training data from Liberty files, removing the need for costly labeling; and (3) an attention-based model architecture that accommodates various pin counts and enables the creation of property-specific cell and arc embeddings. Experimental results demonstrate that Lib2Vec effectively captures functional and electrical similarities. Moreover, linear algebraic operations on cell vectors reveal meaningful relationships, such as vector(BUF) - vector(INV) + vector(NAND) ~ vector(AND), showcasing the framework's nuanced representation capabilities. Lib2Vec also enhances downstream circuit learning applications, especially when labeled data is scarce.
format Preprint
id arxiv_https___arxiv_org_abs_2503_22900
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Library Cell Representations in Vector Space
Liang, Rongjian
Lu, Yi-Chen
Liu, Wen-Hao
Ren, Haoxing
Machine Learning
Hardware Architecture
We propose Lib2Vec, a novel self-supervised framework to efficiently learn meaningful vector representations of library cells, enabling ML models to capture essential cell semantics. The framework comprises three key components: (1) an automated method for generating regularity tests to quantitatively evaluate how well cell representations reflect inter-cell relationships; (2) a self-supervised learning scheme that systematically extracts training data from Liberty files, removing the need for costly labeling; and (3) an attention-based model architecture that accommodates various pin counts and enables the creation of property-specific cell and arc embeddings. Experimental results demonstrate that Lib2Vec effectively captures functional and electrical similarities. Moreover, linear algebraic operations on cell vectors reveal meaningful relationships, such as vector(BUF) - vector(INV) + vector(NAND) ~ vector(AND), showcasing the framework's nuanced representation capabilities. Lib2Vec also enhances downstream circuit learning applications, especially when labeled data is scarce.
title Learning Library Cell Representations in Vector Space
topic Machine Learning
Hardware Architecture
url https://arxiv.org/abs/2503.22900