Machine-learning-based hotspot detection using topological classification and critical feature extraction

Authors:
Yen-Ting Yu;Geng-He Lin;Iris Hui-Ru Jiang;Charles Chiang
Affiliations:
National Chiao-Tung University, Hsin-Chu, Taiwan;National Chiao-Tung University, Hsin-Chu, Taiwan;National Chiao-Tung University, Hsin-Chu, Taiwan;Synopsys, Inc., Mountain View, CA
Venue:
Proceedings of the 50th Annual Design Automation Conference
Year:
2013

Citing 10
Cited 0

A training algorithm for optimal margin classifiers

COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
Support-Vector Networks

Machine Learning
Efficient process-hotspot detection using range pattern matching

Proceedings of the 2006 IEEE/ACM international conference on Computer-aided design
Accurate detection for process-hotspots with vias and incomplete specification

Proceedings of the 2007 IEEE/ACM international conference on Computer-aided design
High performance lithographic hotspot detection using hierarchically refined machine learning

Proceedings of the 16th Asia and South Pacific Design Automation Conference
Rapid layout pattern classification

Proceedings of the 16th Asia and South Pacific Design Automation Conference
LIBSVM: A library for support vector machines

ACM Transactions on Intelligent Systems and Technology (TIST)
Accurate process-hotspot detection using critical design rule extraction

Proceedings of the 49th Annual Design Automation Conference
Improved tangent space based distance metric for accurate lithographic hotspot classification

Proceedings of the 49th Annual Design Automation Conference
ICCAD-2012 CAD contest in fuzzy pattern matching for physical verification and benchmark suite

Proceedings of the International Conference on Computer-Aided Design

Quantified Score

Hi-index	0.00

Visualization

Abstract

Because of the widening sub-wavelength lithography gap in advanced fabrication technology, lithography hotspot detection has become an essential task in design for manufacturability. Current state-of-the-art works unite pattern matching and machine learning engines. Unlike them, we fully exploit the strengths of machine learning using novel techniques. By combing topological classification and critical feature extraction, our hotspot detection framework achieves very high accuracy. Furthermore, to speed up the evaluation, we verify only possible layout clips instead of full-layout scanning. After detection, we filter hotspots to reduce the false alarm. Experimental results show that the proposed framework is very accurate and demonstrates a rapid training convergence. Moreover, our framework outperforms the 2012 CAD Contest at ICCAD winner on accuracy and false alarm.