CiteSeerX — Genetic Programming and Incremental Approaches to Solve Supervised Learning Problems

Genetic Programming and Incremental Approaches to Solve Supervised Learning Problems

Popular Tags

No tags have been applied to this document.

BibTeX | Add To MetaCart

@MISC{Kuscu_geneticprogramming,
    author = {Ibrahim Kuscu},
    title = {Genetic Programming and Incremental Approaches to Solve Supervised Learning Problems},
    year = {}
}

Bookmarks

OpenURL

Abstract:

This paper presents an evolutionary approach and an incremental approach to find learning rules of several supervised learning tasks. In evolutionary approach potential solutions are represented as variable length mathematical (LISP S-) expressions. Thus, it is similar to Genetic Programming (GP) but it employs only a fixed set of non-problem specific functions to solve a variety of problems. The model is tested on three Monks' and parity problems. The results indicate the usefulness of the encoding schema in discovering learning rules for simple supervised learning problems. However, hard learning problems require special attention in terms of their need for larger size codings of the potential solutions and their ability of generalisation over the testing set. In order to find better solutions to these issues, a hill climbing strategy with an incremental coding of potential solutions is used in discovering learning rules for the same problems. It is found that with this strategy larg...

Citations

2044	Learning internal representations by error propagation – Rumelhart, G, et al. - 1986
150	The MONK's problems - a performance comparison of different learning algorithm – Thrun - 1991
53	Genetic Programming II – Koza - 1994
51	Trading spaces: Computation, representation, and the limits of uninformed learning – Clark, Thornton - 1997
20	Genetic Programming:On the Programming of Computers by Means of Natural Selection – Koza - 1992
11	Design of Artificial Neural Networks Using Genetic Algorithms: review and prospect – Kuscu, Thornton - 1994
8	Supervised learning of conditional approach: a case study – Thornton - 1993
8	Measuring the difficulty of specific learning problems – Thornton - 1995
4	Evolution of learning rules for supervised tasks i: Simple learning problems – Kuscu - 1995
4	The genitor algorithm and why rank based-based allocation of reproductive trials is best – Whitley - 1989
3	Incrementally learning the rules for supervised tasks: Monk's problems – Kuscu - 1995
2	Evolution of learning rules for hard learning problems – Kuscu - 1996