基于主动学习和视觉状态空间模型的热点检测器

王盈; 蔡述庭; 熊晓明

doi:10.12052/gdutxb.240114

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于主动学习和视觉状态空间模型的热点检测器

集成电路科学与工程 | 更新时间：2026-05-27

- 基于主动学习和视觉状态空间模型的热点检测器
- A Hotspot Detector Based on Active Learning and Visual State Space Models
- 广东工业大学学报 2024年41卷第6期页码：45-51
- 作者机构：
  
  1. 广东工业大学计算机学院,广东,广州,510006
  2. 广东工业大学集成电路学院,广东,广州,510006
- 作者简介：
  
  王盈(2001–) ，女，硕士研究生，主要研究方向为人工智能辅助电子设计自动化，E-mail：2112205302@mail2.gdut.edu.cn
  熊晓明(1959–) ，男，教授，主要研究方向为超大规模集成电路设计等，E-mail：xmxiong@gdut.edu.cn
- 基金信息：
  
  广东省重点领域研发计划资助项目(2022B0701180001)
- DOI：10.12052/gdutxb.240114
  中图分类号： TN407;TN707
- 纸质出版：2024
- 稿件说明：
移动端阅览
王盈, 蔡述庭, 熊晓明. 基于主动学习和视觉状态空间模型的热点检测器[J]. 广东工业大学学报, 2024, 41(6): 45-51.

Wang Ying, Cai Shu-ting, Xiong Xiao-ming. A Hotspot Detector Based on Active Learning and Visual State Space Models[J]. 2024, 41(6): 45-51.
王盈, 蔡述庭, 熊晓明. 基于主动学习和视觉状态空间模型的热点检测器[J]. 广东工业大学学报, 2024, 41(6): 45-51. DOI： 10.12052/gdutxb.240114.

Wang Ying, Cai Shu-ting, Xiong Xiao-ming. A Hotspot Detector Based on Active Learning and Visual State Space Models[J]. 2024, 41(6): 45-51. DOI： 10.12052/gdutxb.240114.

摘要

物理验证是芯片生产制造中的关键问题，可保证芯片良率。在实际制造前检测芯片版图中的潜在热点是物理验证的重要步骤，确保制造可行性，提高生产效率。传统的热点检测技术具有检测周期长、消耗大量计算资源等问题，不仅增加了整个生产周期的时间成本，而且检测到的热点模式有限。基于主动学习技术和视觉状态空间模型，本文提出一种新的热点检测模型，使用记忆性评估查询的采样策略，缓解热点数据和非热点数据不平衡问题对模型的影响；同时对基于卷积神经网络(Convolutional Neural Network，CNN)结构的分辨率受限以及基于视觉转换器(Vision Transformers，ViT) 网络架构的二次复杂度进行优化，实现热点检测器的线性复杂度。使用ICCAD-2012竞赛数据进行测试，表明本文提出的热点检测器能够显著减少误报率，当召回率高达98.89%时，误报率仅为1.47%。

Abstract

Physical verification is a critical concern in chip manufacturing

ensuring chip yield. Detecting potential hotspots in the chip layout before actual manufacturing is a critical step

which ensures manufacturing feasibility and enhances production efficiency. Traditional hotspot detection techniques suffer from long detection cycles and high computational resource consumption

resulting in increased time costs throughout the production cycle and limited detection of hotspot patterns. Based on active learning techniques and visual state space models

this paper proposes a new hotspot detection model. A memory-based sampling strategy is employed for query evaluation to mitigate the impact of the imbalance between hotspot and non-hotspot data on the model. Furthermore

the resolution constraints of the CNN structure and the secondary complexity of the ViT network architecture are optimized

leading to linear complexity for the hotspot detector. Testing results on the ICCAD-2012 competition dataset show that the proposed hotspot detector significantly reduces the false positive rate

achieving a rate of only 1.47%

while the recall reaches an impressive 98.89%.

关键词

Keywords

references

PAN D Z, YU B, GAO J R. Design for manufacturing with emerging nanolithography[J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2013, 32(10): 1453-1472.

REN P Z, XIAO Y, CHANG X J, et al . A survey of deep active learning[J ] . Association for Computing Machinery, 2021, 54(9): 1-40.

GUPTA P, KAHNG A B, NAKAGAWA S, et al . Lithography simulation-based full-chip design analyses[C ] //Design and Process Integration for Microelectronic Manufacturing IV. San Jose: SPIE, 2006: 277-284.

KIM J, FAN M. Hotspot detection on post-OPC layout using full-chip simulation-based verification tool: a case study with aerial image simulation[C] //23rd Annual BACUS Symposium on Photomask Technology. Monterey: SPIE, 2003: 919-925.

TSENG S S-E, CHANG W C, JIANG I H-R, et al . Efficient search of layout hotspot patterns for matching SEM images using multilevel pixelation[C ] //Optical Microlithography XXXII. San Jose: SPIE, 2019: 51-58.

WEN W Y, LI J C, LIN S Y, et al . A fuzzy-matching model with grid reduction for lithography hotspot detection[J ] . IEEE Transactions on Computer-Aided Design of Integrated Circ uits and Systems, 2014, 33(11): 1671-1680.

YU Y T, CHAN Y C, SINHA S, et al . Accurate process-hotspot detection using critical design rule extraction[C ] //Proceedings of the 49th Annual Design Automation Conference. New York: ACM, 2012: 1167-1172.

GENG H, YANG H Y, ZHANG L, et al . Hotspot detection via attention-based deep layout metric learning[C ] //Proceedings of the 39th International Conference on Computer-Aided Design. New York: ACM, 2020: 1-8.

JIANG Y Y, YANG F, ZHU H L, et al . Efficient layout hotspot detection via binarized residual neural network[C ] //Proceedings of the 56th Annual Design Automation Conference 2019. New York: ACM, 2019: 1-6.

SUN S Y, JIANG Y Y, YANG F, et al . Efficient hotspot detection via graph neural network[C ] //2022 Design, Automation Test in Europe Conference Exhibition (DATE) . Antwerp: IEEE, 2022: 1233-1238.

GAL Y, GHAHRAMANI Z. Dropout as a bayesian approximation: representing model uncertainty in deep learning[C] //international conference on machine learning. New York: PMLR, 2016: 1050-1059.

LIN J H, Divergence measures based on the Shannon entropy[J]. IEEE Transactions on Information theory, 1991, 37(1): 145-151.

TORRES J A. ICCAD-2012 CAD contest in fuzzy pattern matching for physical verification and benchmark suite[C] //Proceedings of the International Conference on Computer-Aided Design. New York: ACM, 2012: 349-350.

YANG H Y, SU J, ZOU Y, et al . Layout hotspot detection with feature tensor generation and deep biased learning[C ] //Proceedings of the 54th Annual Design Automation Conference 2017. New York: ACM, 2017: 1-6.

CHEN Y Y, WU Y W, WANG J Y, et al . LLM-HD: layout language model for hotspot detection with GDS semantic encoding[C ] //Proceedings of the 61th Annual Design Automation Conference 2024. New York: ACM, 2024: 1-6.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向电力场景图像的动态自适应增量目标检测DA-IOD算法

基于域不变判别特征学习的深子域自适应方法

基于深度学习的自适应采样及记忆增强压缩感知方法

多尺度对比嵌入增强的自适应入侵检测模型

面向灾后救援的多无人机通感算一体化资源分配优化方法