Metrics for characterizing machine learning-based hotspot detection methods

Machine learning techniques have recently been applied to the problem of lithographic hotspot detection. It is widely believed that they are capable of identifying hotspot patterns unknown to the trained model. The quality of a machine learning method is conventionally measured by the accuracy rates determined from experiments employing random partitioning of benchmark samples into training and testing sets. In this paper, we demonstrate that these accuracy rates may not reflect the predictive capability of a method. We introduce two metrics - the predictive and memorizing accuracy rates - that quantitatively characterize the method's capability to capture hotspots. We also claim that the number of false alarms per detected hotspot reflects both the method's performance and the difficulty of detecting hotspots in the test set. By adopting the proposed metrics, a designer can conduct a fair comparison between different hotspot detection tools and adopt the one better suited to the verification needs.

[1]  Fedor G. Pikus,et al.  High performance lithographic hotspot detection using hierarchically refined machine learning , 2011, 16th Asia and South Pacific Design Automation Conference (ASP-DAC 2011).

[2]  Jie Yang,et al.  DRC Plus: augmenting standard DRC with pattern matching on 2D geometries , 2007, SPIE Advanced Lithography.

[3]  Fedor G. Pikus,et al.  High Performance Lithography Hotspot Detection with Hierarchically Refined Machine Learning Methods , 2010 .

[4]  Yici Cai,et al.  Efficient range pattern matching algorithm for process-hotspot detection , 2008, IET Circuits Devices Syst..

[5]  Juhwan Kim,et al.  Hotspot detection on post-OPC layout using full-chip simulation-based verification tool: a case study with aerial image simulation , 2003, SPIE Photomask Technology.

[6]  Philippe Hurat,et al.  Layout printability optimization using a silicon simulation methodology , 2004, International Symposium on Signals, Circuits and Systems. Proceedings, SCS 2003. (Cat. No.03EX720).

[7]  Malgorzata Marek-Sadowska,et al.  Rapid layout pattern classification , 2011, 16th Asia and South Pacific Design Automation Conference (ASP-DAC 2011).

[8]  Frank Liu,et al.  Predicting variability in nanoscale lithography processes , 2009, 2009 46th ACM/IEEE Design Automation Conference.

[9]  David Z. Pan,et al.  Synergistic physical synthesis for manufacturability and variability in 45nm designs and beyond , 2008, 2008 Asia and South Pacific Design Automation Conference.

[10]  Reinhard März,et al.  ORC and LfD as first steps towards DfM , 2006, European Mask and Lithography Conference.

[11]  Costas J. Spanos,et al.  Clustering and pattern matching for an automatic hotspot classification and detection system , 2009, Advanced Lithography.

[12]  Malgorzata Marek-Sadowska,et al.  Detecting context sensitive hot spots in standard cell libraries , 2009, Advanced Lithography.

[13]  Puneet Gupta,et al.  Manufacturing-aware physical design , 2003, ICCAD-2003. International Conference on Computer Aided Design (IEEE Cat. No.03CH37486).

[14]  David Z. Pan,et al.  Machine learning based lithographic hotspot detection with critical-feature extraction and classification , 2009, 2009 IEEE International Conference on IC Design and Technology.