Variation-Aware Circuit Macromodeling and Design Based on Surrogate Models - Simulation and Modeling Methodologies, Technologies and Application - page 257

The Gaussian-based Kriging model can be used as an approximation method since
this model is able to provide estimation of the uncertainty in the prediction. Adaptive
sampling methods (e.g. [3]) can be used to balance exploration (improving the general
accuracy of the surrogate model) and exploitation (improving the accuracy of the
surrogate model in the local optimum area) during optimization. An alternative me-
thod, space mapping [25], maps the input/output space of a low-fidelity model to the
input/output space of the high-fidelity model. These methods can significantly im-
prove the optimization efficiency when physically-based and computational efficient
low-fidelity models are available.

6 Summary

This work presents the applications of surrogate modeling in variation-aware circuit
macromodeling and design analysis. Surrogate modeling can facilitate the design
exploration and optimization with variation-aware performance models. Also, surro-
gate modeling can be used to enhance the accuracy and scalability of IO macromo-
dels. Moreover, the surrogate model-based method is able to generate device models
with critical variability parameters. The surrogate-based method greatly reduces the
complexities and costs of variation-aware macromodeling and circuit design.

Acknowledgements. This material is based up on the work supported by the Self-
HEALing mixed signal Integrated Circuits (HEALICs) program of the Department of
Defense Advanced Research Projects Agency (DARPA) and AFRL under contract
number: FA8650-09-C-7925. Approved for Public Release. Distribution Unlimited.
The views expressed are those of the authors and do not reflect the official policy or
position of the Department of Defense or the U.S. Government. The authors T. Zhu,
Dr. P. D. Franzon, and Dr. M. B. Steer would like to thank Dr. T. Dhaene of Ghent
University, Belgium , for providing the SUrrogate MOdeling (SUMO) Toolbox and
for helpful discussion.

References

1. Gorissen, D., Turck, F.D., Dhaene, T.: Evolutionary Model Type Selection for Global Sur-
rogate Modeling. Journal of Machine Learning Research 10(1), 2039-2078 (2009)
2. Eldred, M.S., Dunlavy, D.M.: Formulations for Surrogate-based Optimization with Data
Fit, Multifidelity, and Reduced-order models. In: 11th AIAA/ISSMO Multidisciplinary
Analysis and Optimization Conference, AIAA-2006-7117, Protsmouth, Virginia (2006)
3. Forrester, A., Sobester, A., Keane, A.: Engineering Design via Surrogate Moldeling: A
Practical Guide. John Wiley & Sons (2008)
4. Kleijnen, J.: Design and Analysis of Simulation Experiments. Springer (2008)
5. Crombecq, K., Couckuyt, I., Gorissen, D., Dhaene, T.: Space-Filling Sequential Design
Strategies for Adaptive Surrogate Modelling. In: 1st International Conference on Soft
Computing Technology in Civil, Structural and Environmental Engineering, Paper 50,
Civil-Comp Press, Stirlingshire (2009)
6. Meckesheimer, M., Booker, A.J., Barton, R., Simpson, T.: Computationally Inexpensive
Metamodel Assessment Strategies. AIAA Journal 40(10), 2053-2060 (2002)