Fast Statistical Analysis Using Machine Learning

In this chapter, we describe a fast statistical yield analysis methodology for memory design. At the heart of its engine is a mixture importance sampling-based methodology which comprises a uniform sampling stage and an importance sampling stage. Logistic regression-based machine learning techniques are employed for modeling the circuit response and speeding up the importance sample points simulations. To avoid overfitting, we rely on a cross-validation-based regularization framework for ordered feature selection. The methodology is comprehensive and computationally efficient. We demonstrate the methodology on an industrial state-of-the-art 14 nm FinFET SRAM design with write-assist circuitry. The results corroborate well with hardware and with the fully circuit-simulation-based approach.

[1]  Aleksandra Mojsilovic.  A logistic regression model for small sample classification problems with hidden variables and non-linear relationships: an application in business analytics , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[2]  Rob A. Rutenbar,et al.  Statistical blockade: a novel method for very fast Monte Carlo simulation of rare circuit events, and its application , 2007 .

[3]  Xin Li,et al.  Finding Deterministic Solution From Underdetermined Equation: Large-Scale Performance Variability Modeling of Analog/RF Circuits , 2010, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[4]  Honglak Lee,et al.  Efficient L1 Regularized Logistic Regression , 2006, AAAI.

[5]  James Theiler,et al.  Online Feature Selection using Grafting , 2003, ICML.

[6]  Rajiv V. Joshi,et al.  Sparse Regression Driven Mixture Importance Sampling for Memory Design , 2018, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[7]  Rajiv V. Joshi,et al.  14nm FinFET based supply voltage boosting techniques for extreme low Vmin operation , 2015, 2015 Symposium on VLSI Circuits (VLSI Circuits).

[8]  Rajiv V. Joshi,et al.  Accelerated statistical simulation via on-demand Hermite spline interpolations , 2011, 2011 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[9]  J. S. Hunter,et al.  Statistics for Experimenters: Design, Innovation, and Discovery , 2006 .

[10]  Kaushik Roy,et al.  Statistical design and optimization of SRAM cell for yield enhancement , 2004, ICCAD 2004.

[11]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[12]  Rajiv V. Joshi,et al.  Programmable supply boosting techniques for near threshold and wide operating voltage SRAM , 2017, 2017 IEEE Custom Integrated Circuits Conference (CICC).

[13]  J. F. Ramaley.  Buffon's Noodle Problem , 1969 .

[14]  Randall S. Collica.  A logistic regression yield model for SRAM bit fail patterns , 1993, Proceedings of 1993 IEEE International Workshop on Defect and Fault Tolerance in VLSI Systems.

[15]  D. Hosmer,et al.  Applied Logistic Regression , 1991 .

[16]  N. Metropolis,et al.  The Monte Carlo method. , 1949 .

[17]  Rajan Chattamvelli.  Data Mining Methods , 2009 .

[18]  Sani R. Nassif,et al.  Statistical analysis of SRAM cell stability , 2006, 2006 43rd ACM/IEEE Design Automation Conference.

[19]  Rouwaida Kanj,et al.  Statistical Exploration of the Dual Supply Voltage Space of a 65nm PD/SOI CMOS SRAM Cell , 2006, 2006 European Solid-State Device Research Conference.

[20]  Daniel T. Larose,et al.  Data Mining Methods and Models: Larose/Data Mining Methods and Models , 2005 .

[21]  Ali Chehab,et al.  Regularized logistic regression for fast importance sampling based SRAM yield analysis , 2017, 2017 18th International Symposium on Quality Electronic Design (ISQED).

[22]  Rodney X. Sturdivant,et al.  Applied Logistic Regression: Hosmer/Applied Logistic Regression , 2005 .

[23]  Xin Li,et al.  Statistical regression for efficient high-dimensional modeling of analog and mixed-signal performance variations , 2008, 2008 45th ACM/IEEE Design Automation Conference.

[24]  Christian P. Robert,et al.  Monte Carlo Methods , 2016 .

[25]  T. Minka.  A comparison of numerical optimizers for logistic regression , 2004 .

[26]  Rajiv V. Joshi,et al.  Mixture importance sampling and its application to the analysis of SRAM designs in the presence of rare failure events , 2006, 2006 43rd ACM/IEEE Design Automation Conference.

[27]  D. Böhning.  Multinomial logistic regression algorithm , 1992 .

[28]  Norazan Mohamed Ramli,et al.  Outlier detection in logistic regression and its application in medical data analysis , 2012, 2012 IEEE Colloquium on Humanities, Science and Engineering (CHUSER).