Empirical Analysis of GP Tree-Fragments

Smart, Will; Andreae, Peter; Zhang, Mengjie

doi:10.1007/978-3-540-71605-1_6

Cited by 13 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An alternative way to tackle the huge number of schemata processed by GP proposed by Smart et al is to consider so called maximal fragments [61]. In the authors notation a fragment closely resembles Langdon-style [44] i.e.…”

Section: Offline Methodsmentioning

confidence: 99%

Towards identifying salient patterns in genetic programming individuals

Joó

Neirotti

2009

Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

This thesis addresses the problem of offline identification of salient patterns in genetic programming individuals. It discusses the main issues related to automatic pattern identification systems, namely that these (a) should help in understanding the final solutions of the evolutionary run, (b) should give insight into the course of evolution and (c) should be helpful in optimizing future runs. Moreover, it proposes an algorithm, Extended Pattern Growing Algorithm ([E]PGA) to extract, filter and sort the identified patterns so that these fulfill as many as possible of the following criteria: (a) they are representative for the evolutionary run and/or search space, (b) they are human-friendly and (c) their numbers are within reasonable limits. The results are demonstrated on six problems from different domains.

show abstract

Section: Offline Methodsmentioning

confidence: 99%

Towards identifying salient patterns in genetic programming individuals

Joó

Neirotti

2009

Proceedings of the 11th Annual Conference on Genetic and Evolutionary Computation

View full text Add to dashboard Cite

show abstract

“…Majeed [20] extracted schemata as subtrees with acceptable sizes that occur in at least half of the population and present in the last generation. Smart and Zhang [17] also studied the number and size of fragments in population, the popularity of most frequent fragment and the number of "maximal fragments" over different generations. Some other researchers tried to employ schema theory for improving the GP performance [7,8,34].…”

Section: Related Workmentioning

confidence: 98%

“…Thus, all of previously mentioned schema theories suffer from issues enumerated in Section 1. Except in very few researches [17,20,25], no extraction method is specified for finding the present schemata in a given population by studies of Table 1. Again, it is indicated in the table that researchers rarely provided experimental results for their schema theories.…”

Section: Related Workmentioning

confidence: 99%

“…Again, it is indicated in the table that researchers rarely provided experimental results for their schema theories. Some work analyzed and tracked the schema frequency or other schema theory terms in evolution [9,17,20,21,28]. Poli and Langdon [21] analyzed the empirical aspects of their schema [3] in a separate study.…”

Section: Related Workmentioning

confidence: 99%

“…The most popular definition of a schema in GP literature is "a set of points in the search space that share some syntactic characteristics" [15][16][17]. Considering the principal objective of schema theory that is modeling the behavior of evolutionary algorithms, the schema must be a set of points of the search space which have common behavior.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Semantic schema theory for genetic programming

Zojaji

Ebadzadeh

2015

Appl Intell

View full text Add to dashboard Cite

Schema theory is the most well-known model of evolutionary algorithms. Imitating from genetic algorithms (GA), nearly all schemata defined for genetic programming (GP) refer to a set of points in the search space that share some syntactic characteristics. In GP, syntactically similar individuals do not necessarily have similar semantics. The instances of a syntactic schema do not behave similarly, hence the corresponding schema theory becomes unreliable. Therefore, these theories have been rarely used to improve the performance of GP. The main objective of this study is to propose a schema theory which could be a more realistic model for GP and could be potentially employed for improving GP in practice. To achieve this aim, the concept of semantic schema is introduced. This schema partitions the search space according to semantics of trees, regardless of their syntactic variety. We interpret the semantics of a tree in terms of the mutual information between its output and the target. The semantic schema is characterized by a set of semantic building blocks and their joint probability distribution. After introducing the semantic building blocks, an algorithm for finding them in a given population is presented. An extraction method that looks for the most significant schema of the population is provided. Moreover, an exact microscopic schema theorem is suggested that predicts the expected number of schema samples in the next generation. Experimental results demonstrate the capability of the proposed schema definition in representing the semantics of the schema instances. It is also revealed that the semantic schema theorem estimation is more realistic than previously defined schemata.

show abstract