About the GPEMjournal blog

This is the editor's blog for the journal Genetic Programming and Evolvable Machines. The official web site for the journal, maintained by the publisher (Springer) is here. The GPEMjournal blog is authored and maintained by Lee Spector.

Thursday, February 24, 2011

CFP: Special Issue on Systems Identification

Guest Editors:

Steven Gustafson
GE Global Research, USA
steven D0T gustafson AT research D0T ge D0T com

Una-May O’Reilly
MIT, USA
unamay AT csail D0T mit D0T edu

Genetic programming is a valuable tool for reverse engineering data. Solutions found by Genetic Programming are in the form of algorithms that can be inspected, model checked, verified, and optimized. While this is possible with other classification methods, the intuitive representations GP employs makes it amenable to systems identification, defined here as: 

Systems Identification: the process of exploring and identifying the variables, coefficients, and
model forms that best or most efficiently represent a system.

A recent example of GP for systems identification can be found in Schmidt and Lipson’s 2009 Science article “Distilling Free-form Natural Laws from Experimental Data” (Schmidt and Lipson 2009). Similarly, scientists at Dow Chemical, University of Antwerp, and Evolved Analytics LLC, have developed flexible and robust GP systems that provide key statistics and visualizations during the evolutionary process to guide the human user (Kotanchek, Vladislavleva, and Smits 2009). In this Special Issue, we would like to bring a focus on this unique but extremely valuable application of GP for Systems Identification. Topics of interest include:

• GP approaches to learn laws of various systems, e.g. biological, mechanical and artificial.

• GP approaches to uncover nonlinear relationships between variables in complex systems

• Scalable GP systems that can handle one or more orders of magnitude more than typical systems to enable more real-world Systems Identification, e.g. financial anomaly detection.

• GP systems that provide an improved understanding of the solutions, from variable interaction to improved confidence bounding, e.g. providing statistics of similar to modern packages like Minitab, Matlab, R.

• Approaches that move GP closer to systems like CART as a way to explore variables, relationships, and data, where users can quickly inspect solutions and modify the system to improve performance and capability.

We encourage all prospective authors to contact the guest editors, at the address below, as early as possible, to indicate your intention to submit a paper to this special issue.

Submission Deadline: September 1, 2011

Acceptance Notification: November 15, 2011

Final Manuscript Deadline: January 15, 2012

References

Kotanchek, M. E., Vladislavleva, E. Y., and Smits, G. F. (2009). Symbolic Regression via GP
as a Discovery Engine: Insights on Outliers and Prototypes. In Riolo, R., O'Reilly, U.-M., and

McConaghy, T. Genetic Programming Theory and Practice VII, pp. 55-72, Springer.
http://www.springerlink.com/content/p508hr96008h61t5/

Schmidt, M., and Lipson, H. (2009). Distilling Free-Form Natural Laws from Experimental
Data. Science 324(5923) pp. 81 - 85.