Genetic Programming Theory and Practice XI (Genetic and Evolutionary Computation)
Format: PDF / Kindle (mobi) / ePub
These contributions, written by the foremost international researchers and practitioners of Genetic Programming (GP), explore the synergy between theoretical and empirical results on real-world problems, producing a comprehensive view of the state of the art in GP. Topics in this volume include: evolutionary constraints, relaxation of selection mechanisms, diversity preservation strategies, flexing fitness evaluation, evolution in dynamic environments, multi-objective and multi-modal selection, foundations of evolvability, evolvable and adaptive evolutionary operators, foundation of injecting expert knowledge in evolutionary search, analysis of problem difficulty and required GP algorithm complexity, foundations in running GP on the cloud – communication, cooperation, flexible implementation, and ensemble methods. Additional focal points for GP symbolic regression are: (1) The need to guarantee convergence to solutions in the function discovery mode; (2) Issues on model validation; (3) The need for model analysis workflows for insight generation based on generated GP solutions – model exploration, visualization, variable selection, dimensionality analysis; (4) Issues in combining different types of data. Readers will discover large-scale, real-world applications of GP to a variety of problem domains via in-depth presentations of the latest and most significant results.
(S22) contains the term y which being the dependent variable in the regression. First, the isolate(true) clause in (S22) keeps any of the champions from the first where clause from reaching the final solution list. While there is nothing wrong with using y during training ( y is used repeatedly during fitness calculation while training), allowing y to appear in the final solution causes problems because y will not be available during testing. So only solutions containing the features x0 thru x99.
Be found at http://cscs.umich.edu/gptp-workshops/ . We thank all the workshop participants for making the workshop an exciting and productive 3 days. In particular we thank the authors, without whose hard work and creative talents neither the workshop nor the book would be possible. We also thank our keynote speakers Dr. Charles Ofria, associate professor at Michigan State University in the Department of Computer Science and the Ecology, Evolutionary Biology, and Behavior Program, director.
Http://linkinghub.elsevier.com/retrieve/pii/S0167739X12002324 Wagner N, Michalewicz Z, Khouja M, McGregor R (2007) Time series forecasting for dynamic environments: the DyFor genetic program model. IEEE Trans Evol Comput 11(4):433–452. doi:10.1109/TEVC.2006.882430CrossRef Yu T, Chen SH, Kuo TW (2004) Discovering financial technical trading rules using genetic programming with lambda abstraction. In: O’Reilly UM, Yu T, Riolo RL, Worzel B (eds) Genetic programming theory and practice II.
Measures Predicted Actual Highest unemployment rate 20.5 23.9 Lowest unemployment rate −0.2 2.7 Out of sample R-Square 0.51 Table 7.7Formula with a polynomial benefit replacement effect RSQ = [0.85] Sign Coefficient Variable T-statistic P-values − 0.296193 benefit_replacement_rate_oecd −4.95 0.0000 + 13.831420 ln(benefit_replacement_rate_oecd) 9.31 0.0000 − 2.942167 ln(active_labour_per_unemployed) −8.19 0.0000 + 4.193654E-08 quart(union_density_vis) 2.68.
Experimental setting. Test Problems Used. The first three considered applications are real life problems in the field of pharmacokinetics. They have been chosen because they have been tackled in several previous studies, in particular using GP (see for instance Archetti et al. (2007)). They consist in predicting the value of three different pharmacokinetic parameters of a set of candidate drug compounds based on their molecular structure. The first pharmacokinetic parameter we consider is human.