W. Langdon, and R. Poli. Proceedings of the 2006 IEEE Congress on Evolutionary
Computation, page 6118--6125. Vancouver, IEEE Press, (6-21 July 2006)
Abstract
Particle swarm optimiser and genetic algorithm
populations are macro-organisms, which perceive their
environment as if filtered via a kernel. The kernel
assimilates each individual's sensory abilities so that
the collective moves using a greedy hill-climbing
strategy. This model is fitted to data collected in
real PSO and GA runs by using genetic programming to
evolve the kernel.
In nature animals tend to live within groups. The
social interactions effectively transform the fitness
selection landscape seen by an isolated individual. In
some cases a group behaves (or even can be said to
think) like a single organism. Kernels provide a lens
which coarse-grains or averages individual senses and
so may help explain joint actions and social
responses.
The original multi-modal problem is smoothed by
convolving it with a problem specific filter designed
by GP. Because populations see the transformed social
fitness landscape, they can pass over local optima. GP
can give a good fit between the predicted behaviour of
the macroscopic organism and the actual runs.
%0 Conference Paper
%1 langdon:2006:CEC
%A Langdon, William B.
%A Poli, Riccardo
%B Proceedings of the 2006 IEEE Congress on Evolutionary
Computation
%C Vancouver
%D 2006
%E Yen, Gary G.
%E Wang, Lipo
%E Bonissone, Piero
%E Lucas, Simon M.
%I IEEE Press
%K PSO, XPS algorithms, genetic programming,
%P 6118--6125
%T Finding Social Landscapes for PSOs via Kernels
%U http://ieeexplore.ieee.org/iel5/11108/35623/01688507.pdf?tp=&arnumber=1688507&isnumber=35623
%X Particle swarm optimiser and genetic algorithm
populations are macro-organisms, which perceive their
environment as if filtered via a kernel. The kernel
assimilates each individual's sensory abilities so that
the collective moves using a greedy hill-climbing
strategy. This model is fitted to data collected in
real PSO and GA runs by using genetic programming to
evolve the kernel.
In nature animals tend to live within groups. The
social interactions effectively transform the fitness
selection landscape seen by an isolated individual. In
some cases a group behaves (or even can be said to
think) like a single organism. Kernels provide a lens
which coarse-grains or averages individual senses and
so may help explain joint actions and social
responses.
The original multi-modal problem is smoothed by
convolving it with a problem specific filter designed
by GP. Because populations see the transformed social
fitness landscape, they can pass over local optima. GP
can give a good fit between the predicted behaviour of
the macroscopic organism and the actual runs.
%@ 0-7803-9487-9
@inproceedings{langdon:2006:CEC,
abstract = {Particle swarm optimiser and genetic algorithm
populations are macro-organisms, which perceive their
environment as if filtered via a kernel. The kernel
assimilates each individual's sensory abilities so that
the collective moves using a greedy hill-climbing
strategy. This model is fitted to data collected in
real PSO and GA runs by using genetic programming to
evolve the kernel.
In nature animals tend to live within groups. The
social interactions effectively transform the fitness
selection landscape seen by an isolated individual. In
some cases a group behaves (or even can be said to
think) like a single organism. Kernels provide a lens
which coarse-grains or averages individual senses and
so may help explain joint actions and social
responses.
The original multi-modal problem is smoothed by
convolving it with a problem specific filter designed
by GP. Because populations see the transformed social
fitness landscape, they can pass over local optima. GP
can give a good fit between the predicted behaviour of
the macroscopic organism and the actual runs.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Vancouver},
author = {Langdon, William B. and Poli, Riccardo},
biburl = {https://www.bibsonomy.org/bibtex/26c3a9b23fa6b432873aca69e0cdac046/brazovayeye},
booktitle = {Proceedings of the 2006 IEEE Congress on Evolutionary
Computation},
editor = {Yen, Gary G. and Wang, Lipo and Bonissone, Piero and Lucas, Simon M.},
interhash = {12a40bd6c414d2406b9b4498d34f5df2},
intrahash = {6c3a9b23fa6b432873aca69e0cdac046},
isbn = {0-7803-9487-9},
keywords = {PSO, XPS algorithms, genetic programming,},
month = {6-21 July},
notes = {WCCI 2006 - A joint meeting of the IEEE, the EPS, and
the IEE.
IEEE Catalog Number: 06TH8846D},
pages = {6118--6125},
publisher = {IEEE Press},
size = {8 pages},
timestamp = {2008-06-19T17:45:05.000+0200},
title = {Finding Social Landscapes for {PSO}s via Kernels},
url = {http://ieeexplore.ieee.org/iel5/11108/35623/01688507.pdf?tp=&arnumber=1688507&isnumber=35623},
year = 2006
}