An investigation of local patterns for estimation of distribution genetic programming

We present an improved estimation of distribution (EDA) genetic programming (GP) algorithm which does not rely upon a prototype tree. Instead of using a prototype tree, Operator-Free Genetic Programming learns the distribution of ancestor node chains, "n-grams", in a fit fraction of each generation's population. It then uses this information, via sampling, to create trees for the next generation. Ancestral n-grams are used because an analysis of a GP run conducted by learning depth first graphical models for each generation indicated their emergence as substructures of conditional dependence. We are able to show that our algorithm, without an operator and a prototype tree, achieves, on average, performance close to conventional tree based crossover GP on the problem we study. Our approach sets a direction for pattern-based EDA GP which off ers better tractability and improvements over GP with operators or EDAs using prototype trees.

[1]  Rafal Salustowicz,et al.  Probabilistic Incremental Program Evolution , 1997, Evolutionary Computation.

[2]  H. Iba,et al.  Estimation of distribution programming based on Bayesian network , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[3]  Hussein A. Abbass,et al.  A Survey of Probabilistic Model Building Genetic Programming , 2006, Scalable Optimization via Probabilistic Modeling.

[4]  Hitoshi Iba,et al.  Estimation of Bayesian Network for Program Generation , 2006 .

[5]  D. Goldberg,et al.  Probabilistic Model Building and Competent Genetic Programming , 2003 .

[6]  Hitoshi Iba,et al.  A Bayesian Network Approach to Program Generation , 2008, IEEE Transactions on Evolutionary Computation.

[7]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[8]  Wolfgang Banzhaf,et al.  The evolution of genetic code in Genetic Programming , 1999 .

[9]  Paulien Hogeweg,et al.  Evolutionary Consequences of Coevolving Targets , 1997, Evolutionary Computation.

[10]  An Exploration of Structure Learning in Bayesian Networks , 2012 .

[11]  Martin Pelikan,et al.  Hierarchical Bayesian optimization algorithm: toward a new generation of evolutionary algorithms , 2010, SICE 2003 Annual Conference (IEEE Cat. No.03TH8734).

[12]  Michael O'Neill,et al.  Semantic Aware Crossover for Genetic Programming: The Case for Real-Valued Function Regression , 2009, EuroGP.

[13]  Malcolm I. Heywood,et al.  Introducing probabilistic adaptive mapping developmental genetic programming with redundant mappings , 2007, Genetic Programming and Evolvable Machines.

[14]  Yin Shan Program distribution estimation with grammar models , 2004 .

[15]  Riccardo Poli,et al.  A Linear Estimation-of-Distribution GP System , 2008, EuroGP.

[16]  Robin Harper,et al.  Spatial co-evolution in Age Layered Planes (SCALP) , 2010, IEEE Congress on Evolutionary Computation.

[17]  Michael O'Neill,et al.  A Fine-Grained View of Phenotypes and Locality in Genetic Programming , 2011 .