Initialization method for grammar-guided genetic programming

García-Arnau, Marc; Manrique, Daniel; Ríos, José; Rodríguez‐Patón, Alfonso

doi:10.1016/j.knosys.2006.11.006

Cited by 25 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous research on the initialization of GP populations has shown that the initialization process can vitally affect the performance of GP algorithms [18,22,9]. However, most of this research has focused on generating a diversity of valid GP tree structures, which may not be useful in all application domains.…”

Section: Model Validationmentioning

confidence: 99%

Identifying and Harnessing the Building Blocks of Machine Learning Pipelines for Sensible Initialization of a Data Science Automation Tool

Olson

Moore

2018

Genetic and Evolutionary Computation

View full text Add to dashboard Cite

As data science continues to grow in popularity, there will be an increasing need to make data science tools more scalable, flexible, and accessible. In particular, automated machine learning (AutoML) systems seek to automate the process of designing and optimizing machine learning pipelines. In this chapter, we present a genetic programming-based AutoML system called TPOT that optimizes a series of feature preprocessors and machine learning models with the goal of maximizing classification accuracy on a supervised classification problem. Further, we analyze a large database of pipelines that were previously used to solve various supervised classification problems and identify 100 short series of machine learning operations that appear the most frequently, which we call the building blocks of machine learning pipelines. We harness these building blocks to initialize TPOT with promising solutions, and find that this sensible initialization method significantly improves TPOT's performance on one benchmark at no cost of significantly degrading performance on the others. Thus, sensible initialization with machine learning pipeline building blocks shows promise for GP-based AutoML systems, and should be further refined in future work.

show abstract

Section: Model Validationmentioning

confidence: 99%

Identifying and Harnessing the Building Blocks of Machine Learning Pipelines for Sensible Initialization of a Data Science Automation Tool

Olson

Moore

2018

Genetic and Evolutionary Computation

View full text Add to dashboard Cite

show abstract

“…A grammar-guided genetic program (GGGP) is an evolutionary system that could potentially find solutions to any problem whose syntactic restrictions can be formally defined by a CFG [5]. Individuals are derivation trees of the CFG that, when the algorithm starts, are generated by a grammar-based initialization method [6]. Neither this method nor crossover and mutation operators can generate invalid individuals because they are not contained in the language described by the CFG [4].…”

Section: The Card Game Languagementioning

confidence: 99%

A Card Game Description Language

Font

Mahlmann

Manrique

et al. 2013

Applications of Evolutionary Computation

View full text Add to dashboard Cite

Abstract. We present initial research regarding a system capable of generating novel card games. We furthermore propose a method for computationally analysing existing games of the same genre. Ultimately, we present a formalisation of card game rules, and a context-free grammar G cardgame capable of expressing the rules of a large variety of card games. Example derivations are given for the poker variant Texas hold 'em, Blackjack and UNO. Stochastic simulations are used both to verify the implementation of these well-known games, and to evaluate the results of new game rules derived from the grammar. In future work, this grammar will be used to evolve completely novel card games using a grammarguided genetic program.

show abstract

“…Still, there is research work going on regarding this issue of finding optimal initialization techniques as it is a fact that the use of different initialization strategies can lead to very different overall results (as for example demonstrated in [HHM04]). For example, there are approaches that produce initial populations that are generated adequately distributed in terms of tree size and distribution within the search space [GAMRRP07].…”

Section: Initializationmentioning

confidence: 99%