Efficient performance modeling via Dual-Prior Bayesian Model Fusion for analog and mixed-signal circuits

In this paper, we propose a novel Dual-Prior Bayesian Model Fusion (DP-BMF) algorithm for performance modeling. Different from the previous BMF methods which use only one source of prior knowledge, DP-BMF takes advantage of multiple sources of prior knowledge to fully exploit the available information and, hence, further reduce the modeling cost. Based on a graphical model, an efficient Bayesian inference is developed to fuse two different prior models and combine the prior information with a small number of training samples to achieve high modeling accuracy. Several circuit examples demonstrate that the proposed method can achieve up to 1.83× cost reduction over the traditional one-prior BMF method without surrendering any accuracy.

[1]  Jean-Olivier Plouchart,et al.  Co-Learning Bayesian Model Fusion: Efficient performance modeling of analog and mixed-signal circuits using side information , 2015, 2015 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[2]  R. Bharat Rao,et al.  Bayesian Co-Training , 2007, J. Mach. Learn. Res..

[3]  Abhijit Chatterjee,et al.  Yield Recovery of RF Transceiver Systems Using Iterative Tuning-Driven Power-Conscious Performance Optimization , 2015, IEEE Design & Test.

[4]  Chenjie Gu,et al.  Bayesian Model Fusion: Large-scale performance modeling of analog and mixed-signal circuits by reusing early-stage data , 2013, 2013 50th ACM/EDAC/IEEE Design Automation Conference (DAC).

[5]  Dongsheng Ma,et al.  Principle Hessian direction based parameter reduction with process variation , 2007, ICCAD 2007.

[6]  Rob A. Rutenbar,et al.  Beyond Low-Order Statistical Response Surfaces: Latent Variable Regression for Efficient, Highly Nonlinear Fitting , 2007, 2007 44th ACM/IEEE Design Automation Conference.

[7]  Janet Roveda,et al.  Principle Hessian Direction-Based Parameter Reduction for Interconnect Networks With Process Variation , 2007, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[8]  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.

[9]  Georges G. E. Gielen,et al.  Template-Free Symbolic Performance Modeling of Analog Circuits via Canonical-Form Functions and Genetic Programming , 2009, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[10]  Chenjie Gu,et al.  Efficient parametric yield estimation of analog/mixed-signal circuits via Bayesian model fusion , 2012, 2012 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[11]  Lawrence T. Pileggi,et al.  Asymptotic Probability Extraction for Nonnormal Performance Distributions , 2007, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[12]  Fan Yang,et al.  Efficient multivariate moment estimation via Bayesian model fusion for analog and mixed-signal circuits , 2015, 2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC).

[13]  Trent McConaghy,et al.  High-dimensional statistical modeling and analysis of custom integrated circuits (invited paper) , 2011, 2011 IEEE Custom Integrated Circuits Conference (CICC).

[14]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[15]  Glen Kramer,et al.  Application-specific worst case corners using response surfaces and statistical models , 2004, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[16]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.