Learning Programs in Different Paradigms using Genetic
Programming
M. Wong, and K. Leung. Proceedings of the Fourth Congress of the Italian
Association for Artificial Intelligence, Springer-Verlag, (1995)
Abstract
Genetic Programming (GP) is a method of automatically
inducing programs by representing them as parse trees.
In theory, programs in any computer languages can be
translated to parse trees. Hence, GP should be able to
handle them as well. In practice, the syntax of Lisp is
so simple and uniform that the translation process can
be achieved easily, programs evolved by GP are usually
expressed in Lisp. This paper presents a flexible
framework that programs in various programming
languages can be acquired. This framework is based on a
formalism of logic grammars. To implement the
framework, a system called LOGENPRO (The LOgic grammar
based GENetic PROgramming system) has been developed.
An experiment that employs LOGENPRO to induce a
S-expression for calculating dot product has been
performed. This experiment illustrates that LOGENPRO,
when used with knowledge of data types, accelerates the
learning of programs. Other experiments have been done
to illustrate the ability of LOGENPRO in inducing
programs in difference programming languages including
Prolog and C. These experiments prove that LOGENPRO is
very flexible.
%0 Conference Paper
%1 wong:1995:lpdpGP
%A Wong, Man Leung
%A Leung, Kwong Sak
%B Proceedings of the Fourth Congress of the Italian
Association for Artificial Intelligence
%D 1995
%I Springer-Verlag
%K algorithms, genetic programming
%T Learning Programs in Different Paradigms using Genetic
Programming
%U http://cptra.ln.edu.hk/~mlwong/conference/aiia1995.pdf
%X Genetic Programming (GP) is a method of automatically
inducing programs by representing them as parse trees.
In theory, programs in any computer languages can be
translated to parse trees. Hence, GP should be able to
handle them as well. In practice, the syntax of Lisp is
so simple and uniform that the translation process can
be achieved easily, programs evolved by GP are usually
expressed in Lisp. This paper presents a flexible
framework that programs in various programming
languages can be acquired. This framework is based on a
formalism of logic grammars. To implement the
framework, a system called LOGENPRO (The LOgic grammar
based GENetic PROgramming system) has been developed.
An experiment that employs LOGENPRO to induce a
S-expression for calculating dot product has been
performed. This experiment illustrates that LOGENPRO,
when used with knowledge of data types, accelerates the
learning of programs. Other experiments have been done
to illustrate the ability of LOGENPRO in inducing
programs in difference programming languages including
Prolog and C. These experiments prove that LOGENPRO is
very flexible.
@inproceedings{wong:1995:lpdpGP,
abstract = {Genetic Programming (GP) is a method of automatically
inducing programs by representing them as parse trees.
In theory, programs in any computer languages can be
translated to parse trees. Hence, GP should be able to
handle them as well. In practice, the syntax of Lisp is
so simple and uniform that the translation process can
be achieved easily, programs evolved by GP are usually
expressed in Lisp. This paper presents a flexible
framework that programs in various programming
languages can be acquired. This framework is based on a
formalism of logic grammars. To implement the
framework, a system called LOGENPRO (The LOgic grammar
based GENetic PROgramming system) has been developed.
An experiment that employs LOGENPRO to induce a
S-expression for calculating dot product has been
performed. This experiment illustrates that LOGENPRO,
when used with knowledge of data types, accelerates the
learning of programs. Other experiments have been done
to illustrate the ability of LOGENPRO in inducing
programs in difference programming languages including
Prolog and C. These experiments prove that LOGENPRO is
very flexible.},
added-at = {2008-06-19T17:35:00.000+0200},
author = {Wong, Man Leung and Leung, Kwong Sak},
biburl = {https://www.bibsonomy.org/bibtex/2c445cca2f0e3a2a2d7f05e62603a5634/brazovayeye},
booktitle = {Proceedings of the Fourth Congress of the Italian
Association for Artificial Intelligence},
interhash = {5308e8244b31580a7f5247df9e9422eb},
intrahash = {c445cca2f0e3a2a2d7f05e62603a5634},
keywords = {algorithms, genetic programming},
publisher = {Springer-Verlag},
publisher_address = {Berlin, Germany},
timestamp = {2008-06-19T17:54:20.000+0200},
title = {Learning Programs in Different Paradigms using Genetic
Programming},
url = {http://cptra.ln.edu.hk/~mlwong/conference/aiia1995.pdf},
year = 1995
}