Softening the Structural Difficulty in Genetic Programming with TAG-Based Representation and Insertion/Deletion Operators

In a series of papers [3-8], Daida et. al. highlighted the difficulties posed to Genetic Programming (GP) by the complexity of the structural search space, and attributed the problem to the expression tree representation in GP. In this paper, we show how to transform a fixed-arity expression tree in GP to a non fixed-arity tree (Catalan tree) using representation based on Tree Adjoining Grammars (TAGs). This non fixed-arity property, which is called feasibility, allows us to design many types of genetic operators (as in [16]). In particular, insertion/deletion operators arising naturally from the representation play a role as structural mutation operators. By using these dual operators on TAG-based representation, we demonstrate how these operators can help to soften the structural search difficulties in GP.

[1]  Wolfgang Banzhaf,et al.  Genetic Programming: An Introduction , 1997 .

[2]  Rachel Harrison,et al.  Characterizing a Tunably Difficult Problem in Genetic Programming , 2000, GECCO.

[3]  Peter A. Whigham,et al.  Grammatical bias for evolutionary learning , 1996 .

[4]  Aravind K. Joshi,et al.  Mathematical and computational aspects of lexicalized grammars , 1990 .

[5]  Justinian P. Rosca.  Proceedings of the Workshop on Genetic Programming: From Theory to Real-World Applications , 1995 .

[6]  Peter A. Whigham,et al.  Grammatically-based Genetic Programming , 1995 .

[7]  Hussein A. Abbass,et al.  Tree Adjoining Grammars, Language Bias, and Genetic Programming , 2003, EuroGP.

[8]  Jason M. Daida,et al.  Challenges with Verification, Repeatability, and Meaningful Comparisons in Genetic Programming , 1997 .

[9]  Conor Ryan,et al.  Grammatical Evolution , 2001, Genetic Programming Series.

[10]  Jason M. Daida,et al.  Limits to expression in genetic programming: lattice-aggregate modeling , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[11]  Peter J. Angeline,et al.  On Using Syntactic Constraints with Genetic Programming , 1996 .

[12]  Peter Nordin,et al.  Genetic programming - An Introduction: On the Automatic Evolution of Computer Programs and Its Applications , 1998 .

[13]  Jason M. Daida,et al.  What Makes a Problem GP-Hard? Validating a Hypothesis of Structural Causes , 2003, GECCO.

[14]  Aravind K. Joshi,et al.  Tree-Adjoining Grammars , 1997, Handbook of Formal Languages.

[15]  Nguyen Xuan Hoai,et al.  Solving the symbolic regression problem with tree-adjunct grammar guided genetic programming: the comparative results , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[16]  Leonardo Vanneschi,et al.  Fitness distance correlation in genetic programming: a constructive counterexample , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[17]  Richard C. Waters,et al.  Tree Insertion Grammar: A Cubic-Time, Parsable Formalism that Lexicalizes Context-Free Grammar without Changing the Trees Produced , 1995, CL.

[18]  Vidroha Debroy,et al.  Genetic Programming , 1998, Lecture Notes in Computer Science.

[19]  Nguyen Xuan Hoai,et al.  An investigation on the roles of insertion and deletion operators in tree adjoining grammar guided genetic programming , 2004, Proceedings of the 2004 Congress on Evolutionary Computation (IEEE Cat. No.04TH8753).

[20]  John R. Koza,et al.  Genetic programming 1997 : proceedings of the Second Annual Conference, July 13-16, 1997, Stanford University , 1997 .

[21]  Stuart M. Shieber,et al.  An Alternative Conception of Tree-Adjoining Derivation , 1992, ACL.

[22]  Grzegorz Rozenberg,et al.  Handbook of Formal Languages , 1997, Springer Berlin Heidelberg.

[23]  Aravind K. Joshi,et al.  A study of tree adjoining grammars , 1987 .

[24]  Aravind K. Joshi,et al.  Tree Adjunct Grammars , 1975, J. Comput. Syst. Sci..

[25]  David J. Weir,et al.  Characterizing mildly context-sensitive grammar formalisms , 1988 .

[26]  Man Leung Wong,et al.  Evolutionary Program Induction Directed by Logic Grammars , 1997, Evolutionary Computation.

[27]  Peter A. Whigham,et al.  Search bias, language bias and genetic programming , 1996 .

[28]  J. K. Kinnear,et al.  Advances in Genetic Programming , 1994 .

[29]  Jason M. Daida,et al.  What Makes a Problem GP-Hard? Analysis of a Tunably Difficult Problem in Genetic Programming , 1999, Genetic Programming and Evolvable Machines.

[30]  Andreas Geyer-Schulz,et al.  Fuzzy Rule-Based Expert Systems and Genetic Machine Learning , 1996 .

[31]  Philippe Flajolet,et al.  An introduction to the analysis of algorithms , 1995 .

[32]  Nguyen Xuan Hoai,et al.  A Framework For Tree-Adjunct Grammar Guided Genetic Programming , 2001 .

[33]  Jason M. Daida,et al.  Identifying Structural Mechanisms in Standard Genetic Programming , 2003, GECCO.

[34]  Jason M. Daida,et al.  Analysis of single-node (building) blocks in genetic programming , 1999 .

[35]  Leonardo Vanneschi,et al.  Fitness Distance Correlation in Structural Mutation Genetic Programming , 2003, EuroGP.