CAFFEINE: Template-Free Symbolic Model Generation of Analog Circuits via Canonical Form Functions and Genetic Programming

McConaghy, Trent; Eeckelaert, Tom; Gielen, Georges

doi:10.1109/date.2005.89

Cited by 35 publications

(20 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used the same test setup as in the original CAFFEINE paper [3]. The circuit being modeled is a high-voltage CMOS OTA as shown in Figure 5.…”

Section: Methodsmentioning

confidence: 99%

“…First, we note that our operators always have arity of 1 or 2, which means exponential growth as depth grows is reasonable. Also, even though the focus application is sizing, maintaining the constraint of "interpretability" from [3] acts as a useful surrogate to reduce overfitting. Using that, we choose never to embed one nonlinear operator into another one.…”

Section: Double-strength Caffeinementioning

confidence: 99%

“…a multi-objective problem). This paper has an additional aim: make CAFFEINE fast enough to be used within the context of optimization; specifically, make it at least 3x faster than [3].…”

Section: Problem Formulationmentioning

confidence: 99%

“…Unlike other approaches, CAFFEINE [3] generates symbolic models with open-ended functional forms, i.e. without a priori supplied templates.…”

Section: Introductionmentioning

confidence: 99%

“…Symbolic modeling approaches [2,3] use SPICE-like simulation data to generate interpretable mathematical expressions that relate the circuit performances to the design variables, i.e. "performance models".…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Double-Strength CAFFEINE: Fast Template-Free Symbolic Modeling of Analog Circuits via Implicit Canonical Form Functions and Explicit Introns

McConaghy

Gielen

2006

Proceedings of the Design Automation &Amp; Test in Europe Conference

Self Cite

View full text Add to dashboard Cite

CAFFEINE, introduced previously, automatically generates nonlinear, template-free symbolic performance models of analog circuits from SPICE data. Its key was a directly-interpretable functional form, found via evolutionary search. In application to automated sizing of analog circuits, CAFFEINE was shown to have the best predictive ability from among 10 regression techniques, but was too slow to be used practically in the optimization loop.In this paper, we describe Double-Strength CAFFEINE, which is designed to be fast enough for automated sizing, yet retain good predictive abilities. We design "smooth, uniform" search operators which have been shown to greatly improve efficiency in other domains. Such operators are not straightforward to design; we achieve them in functions by simultaneously making the grammar-constrained functional form implicit, and embedding explicit 'introns' (subfunctions appearing in the candidate that are not expressed). Experimental results on six test problems show that Double-Strength CAFFEINE achieves an average speedup of 5x on the most challenging problems and 3x overall; thus making the technique fast enough for automated sizing.

show abstract

“…We used the same test setup as in the original CAFFEINE paper [3]. The circuit being modeled is a high-voltage CMOS OTA as shown in Figure 5.…”

Section: Methodsmentioning

confidence: 99%

Section: Double-strength Caffeinementioning

confidence: 99%

“…a multi-objective problem). This paper has an additional aim: make CAFFEINE fast enough to be used within the context of optimization; specifically, make it at least 3x faster than [3].…”

Section: Problem Formulationmentioning

confidence: 99%

“…Unlike other approaches, CAFFEINE [3] generates symbolic models with open-ended functional forms, i.e. without a priori supplied templates.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Double-Strength CAFFEINE: Fast Template-Free Symbolic Modeling of Analog Circuits via Implicit Canonical Form Functions and Explicit Introns

McConaghy

Gielen

2006

Proceedings of the Design Automation &Amp; Test in Europe Conference

Self Cite

View full text Add to dashboard Cite

show abstract

Genetic Programming in Industrial Analog CAD: Applications and Challenges

McConaghy

Gielen

Genetic Programming Theory and Practice III

View full text Add to dashboard Cite

Symbolic Regression on Network Properties

Märtens

Kuipers

Mieghem

2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Networks are continuously growing in complexity, which creates challenges for determining their most important characteristics. While analytical bounds are often too conservative, the computational effort of algorithmic approaches does not scale well with network size. This work uses Cartesian Genetic Programming for symbolic regression to evolve mathematical equations that relate network properties directly to the eigenvalues of network adjacency and Laplacian matrices. In particular, we show that these eigenvalues are powerful features to evolve approximate equations for the network diameter and the isoperimetric number, which are hard to compute algorithmically. Our experiments indicate a good performance of the evolved equations for several realworld networks and we demonstrate how the generalization power can be influenced by the selection of training networks and feature sets.

show abstract

CAFFEINE: Template-Free Symbolic Model Generation of Analog Circuits via Canonical Form Functions and Genetic Programming

Abstract: Abstract

Cited by 35 publications

References 14 publications

Double-Strength CAFFEINE: Fast Template-Free Symbolic Modeling of Analog Circuits via Implicit Canonical Form Functions and Explicit Introns

Double-Strength CAFFEINE: Fast Template-Free Symbolic Modeling of Analog Circuits via Implicit Canonical Form Functions and Explicit Introns

Genetic Programming in Industrial Analog CAD: Applications and Challenges

Symbolic Regression on Network Properties

Contact Info

Product

Resources

About