The Effect of Explicit Communications on the
Generality and Robustness of Evolved Team of Agents in
Predator-Prey Problem
I. Tanev, и M. Brozozowski. Proceedings of The First Asian-Pacific Workshop on
Genetic Programming, стр. 31--37. Rydges (lakeside) Hotel, Canberra, Australia, (8 December 2003)
Аннотация
We present a comparative analysis of the effect of
explicit communications among the predator agents on
the generality and robustness of emerged social
behaviour of agents in predator-prey pursuit problem.
The social behavior is evolved employing strongly typed
genetic programming with exception handling
capabilities. We demonstrated that the relatively
complex, social behavior emerges from simple, basic
model of implicit, proximity defined interactions among
the agents. Such model offers the benefits of
simplicity and scalability. However, compared to the
additionally proposed model of explicit communications
among the agents, it features increased computational
effort and inferior generality and robustness of the
emergent behavior of the agents situated in noisy and
uncertain environments. Explicit communications
contribute to the anomalous reduction of performance
degradation with increase of noise in agents
perceptions. The reason for the anomaly is viewed in
the favourable effect of adaptive, situationally
dependent noise in the indirectly (trough
communications) perceived environment.
%0 Conference Paper
%1 Tanev:aspgp03
%A Tanev, Ivan
%A Brozozowski, Michael
%B Proceedings of The First Asian-Pacific Workshop on
Genetic Programming
%C Rydges (lakeside) Hotel, Canberra, Australia
%D 2003
%E Cho, Sung-Bae
%E Hoai, Nguyen Xuan
%E Shan, Yin
%K STGP algorithms, genetic programming,
%P 31--37
%T The Effect of Explicit Communications on the
Generality and Robustness of Evolved Team of Agents in
Predator-Prey Problem
%X We present a comparative analysis of the effect of
explicit communications among the predator agents on
the generality and robustness of emerged social
behaviour of agents in predator-prey pursuit problem.
The social behavior is evolved employing strongly typed
genetic programming with exception handling
capabilities. We demonstrated that the relatively
complex, social behavior emerges from simple, basic
model of implicit, proximity defined interactions among
the agents. Such model offers the benefits of
simplicity and scalability. However, compared to the
additionally proposed model of explicit communications
among the agents, it features increased computational
effort and inferior generality and robustness of the
emergent behavior of the agents situated in noisy and
uncertain environments. Explicit communications
contribute to the anomalous reduction of performance
degradation with increase of noise in agents
perceptions. The reason for the anomaly is viewed in
the favourable effect of adaptive, situationally
dependent noise in the indirectly (trough
communications) perceived environment.
%@ 0-9751724-0-9
@inproceedings{Tanev:aspgp03,
abstract = {We present a comparative analysis of the effect of
explicit communications among the predator agents on
the generality and robustness of emerged social
behaviour of agents in predator-prey pursuit problem.
The social behavior is evolved employing strongly typed
genetic programming with exception handling
capabilities. We demonstrated that the relatively
complex, social behavior emerges from simple, basic
model of implicit, proximity defined interactions among
the agents. Such model offers the benefits of
simplicity and scalability. However, compared to the
additionally proposed model of explicit communications
among the agents, it features increased computational
effort and inferior generality and robustness of the
emergent behavior of the agents situated in noisy and
uncertain environments. Explicit communications
contribute to the anomalous reduction of performance
degradation with increase of noise in agents
perceptions. The reason for the anomaly is viewed in
the favourable effect of adaptive, situationally
dependent noise in the indirectly (trough
communications) perceived environment.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Rydges (lakeside) Hotel, Canberra, Australia},
author = {Tanev, Ivan and Brozozowski, Michael},
biburl = {https://www.bibsonomy.org/bibtex/21301bd064807e1a59074646c688a5fb3/brazovayeye},
booktitle = {Proceedings of The First Asian-Pacific Workshop on
Genetic Programming},
editor = {Cho, Sung-Bae and Hoai, Nguyen Xuan and Shan, Yin},
interhash = {25202e0f49fa50c52db2cbeaceacb80c},
intrahash = {1301bd064807e1a59074646c688a5fb3},
isbn = {0-9751724-0-9},
keywords = {STGP algorithms, genetic programming,},
month = {8 December},
notes = {aspgp03. XML},
pages = {31--37},
timestamp = {2008-06-19T17:52:37.000+0200},
title = {The Effect of Explicit Communications on the
Generality and Robustness of Evolved Team of Agents in
Predator-Prey Problem},
year = 2003
}