Seeding methods for run transferable libraries

  title={Seeding methods for run transferable libraries},
  author={Gearoid Murphy and Conor Ryan},
  booktitle={GECCO '07},
Run Transferable Libraries (RTL) is an extension for GP where individualsin a population choose functions from an external library of ADF-likefunctions rather than from a set of standard GP functions. All previous work done with RTL provided a predefined function set. Thiswork investigates mechanisms by which the library can be seeded with domainrelevent functionality. . 

Figures from this paper

Improving Module Identification and Use in Grammatical Evolution
Grammar Augmentation through Module Encapsulation (GAME) was tested on seven problems from three different domains and was observed to significantly improve the performance on 3 problems and never showing harmful effects on any problem.
Transfer learning in constructive induction with Genetic Programming
This work studies the performance of transferred solutions when the source and target tasks are from different domains, and when they do not share a similar input feature space, and shows that it is possible to predict the success of TL in some cases, particularly in classification tasks.


Run Transferable Libraries - Learning Functional Bias in Problem Domains
It is demonstrated that a system using Run Transferable Libraries can solve a selection of standard benchmarks considerably more quickly than GP with ADFs by building knowledge about a problem.
Evolving Modules in Genetic Programming by Subtree Encapsulation
This paper investigates the effect of encapsulating tree-based genetic programming subtrees by representing them as atoms in the terminal set, so that the subtree evaluations can be exploited as terminal data.
Genetic Programming II: Automatic Discovery of Reusable Programs.
This book presents evidence that it is possible to interpret GP with ADFs as performing either a top-down process of problem decomposition or a bottom-up process of representational change to exploit identified regularities.
The Evolutionary Induction of Subroutines
A genetic algorithm capable of evolving large programs is described by exploiting two new genetic operators which construct and deconstruct parameterized subroutines which help to solve the scaling problem for a class of genetic problem solving methods.
Evaluating GP schema in context
It is shown that the methodology proposed can be used to identify schemata that are important in terms of both individual runs and individual problems (that is, schema that will be important across many runs on a particular problem).
Discovery of subroutines in genetic programming
This chapter discusses an extension of the genetic programming (GP) paradigm based on the idea that subroutines obtained from blocks of good representations act as building blocks and may enable a faster evolution of even better representations.
A Sequential Niche Technique for Multimodal Function Optimization
An algorithm based on a traditional genetic algorithm that involves iterating the GA but uses knowledge gained during one iteration to avoid re-searching, on subsequent iterations, regions of problem space where solutions have already been found.
Iterative Refinement Of Computational Circuits Using Genetic Programming
This paper presents a set of results showing the effectiveness of this approach over multiple iterations for generating squaring, square root, and cubing computational circuits, and performs iterative refinement on a recently patented cubic signal generator circuit, obtaining a refined circuit that is 7.2 times more accurate than the original patented circuit.
Evolutionary algorithms and emergent intelligence
Evolutionary algorithms, computations that are modeled after natural selection are analyzed and proposed to be a novel form of weak method that provide an ideal medium for implementing emergent intelligence.
Alternatives in automatic function definition: a comparison of performance