R. Poli. New Ideas in Optimization, chapter 27, McGraw-Hill, Maidenhead, Berkshire, England, (1999)
Abstract
This chapter describes Parallel Distributed Genetic
Programming (PDGP), a form of Genetic Programming (GP)
which is suitable for the development of programs with
a high degree of parallelism and an efficient and
effective reuse of partial results. Programs are
represented in PDGP as graphs with nodes representing
functions and terminals, and links representing the
flow of control and results. In the simplest form of
PDGP links are directed and unlabelled, in which case
PDGP can be considered a generalisation of standard GP.
However, more complex representations can be used,
which allow the exploration of a large space of
possible programs including standard tree-like
programs, logic networks, neural networks, recurrent
transition networks, finite state automata, etc. In
PDGP, programs are manipulated by special crossover and
mutation operators which guarantee the syntactic
correctness of the offspring. For this reason PDGP
search is very efficient. PDGP programs can be
execut...
This is the most complete account of PDGP and its
performance so far. E.g. a symbolic regression problem
x^6-2*x^4+x^2 in which PDGP does 16 times better than
std GP and 13 times better than GP with ADFs.
XOR, lawnmower, sextic polynomial, encoder-decoder FSA
induction, NLP,
%0 Book Section
%1 Poli:1999:nio
%A Poli, Riccardo
%B New Ideas in Optimization
%C Maidenhead, Berkshire, England
%D 1999
%E Corne, David
%E Dorigo, Marco
%E Glover, Fred
%I McGraw-Hill
%K PDGP algorithms, genetic programming,
%P 403--431
%T Parallel Distributed Genetic Programming
%U http://cswww.essex.ac.uk/staff/rpoli/papers/Poli-NIO-1999-PDGP.pdf
%X This chapter describes Parallel Distributed Genetic
Programming (PDGP), a form of Genetic Programming (GP)
which is suitable for the development of programs with
a high degree of parallelism and an efficient and
effective reuse of partial results. Programs are
represented in PDGP as graphs with nodes representing
functions and terminals, and links representing the
flow of control and results. In the simplest form of
PDGP links are directed and unlabelled, in which case
PDGP can be considered a generalisation of standard GP.
However, more complex representations can be used,
which allow the exploration of a large space of
possible programs including standard tree-like
programs, logic networks, neural networks, recurrent
transition networks, finite state automata, etc. In
PDGP, programs are manipulated by special crossover and
mutation operators which guarantee the syntactic
correctness of the offspring. For this reason PDGP
search is very efficient. PDGP programs can be
execut...
%& 27
%@ 0-07-709506-5
@incollection{Poli:1999:nio,
abstract = {This chapter describes Parallel Distributed Genetic
Programming (PDGP), a form of Genetic Programming (GP)
which is suitable for the development of programs with
a high degree of parallelism and an efficient and
effective reuse of partial results. Programs are
represented in PDGP as graphs with nodes representing
functions and terminals, and links representing the
flow of control and results. In the simplest form of
PDGP links are directed and unlabelled, in which case
PDGP can be considered a generalisation of standard GP.
However, more complex representations can be used,
which allow the exploration of a large space of
possible programs including standard tree-like
programs, logic networks, neural networks, recurrent
transition networks, finite state automata, etc. In
PDGP, programs are manipulated by special crossover and
mutation operators which guarantee the syntactic
correctness of the offspring. For this reason PDGP
search is very efficient. PDGP programs can be
execut...},
added-at = {2008-06-19T17:46:40.000+0200},
address = {Maidenhead, Berkshire, England},
author = {Poli, Riccardo},
biburl = {https://www.bibsonomy.org/bibtex/2ca271303b36693938a5637e56db02e50/brazovayeye},
booktitle = {New Ideas in Optimization},
chapter = 27,
editor = {Corne, David and Dorigo, Marco and Glover, Fred},
interhash = {4bcac060faf32050ff612c2729d03bda},
intrahash = {ca271303b36693938a5637e56db02e50},
isbn = {0-07-709506-5},
keywords = {PDGP algorithms, genetic programming,},
notes = {This is the most complete account of PDGP and its
performance so far. E.g. a symbolic regression problem
x^6-2*x^4+x^2 in which PDGP does 16 times better than
std GP and 13 times better than GP with ADFs.
XOR, lawnmower, sextic polynomial, encoder-decoder FSA
induction, NLP,},
pages = {403--431},
publisher = {McGraw-Hill},
series = {Advanced Topics in Computer Science},
size = {29 pages},
timestamp = {2008-06-19T17:49:43.000+0200},
title = {Parallel Distributed Genetic Programming},
url = {http://cswww.essex.ac.uk/staff/rpoli/papers/Poli-NIO-1999-PDGP.pdf},
year = 1999
}