Authors
Masao [email protected]
Department of Computer Science, National Defense Academy, Hashirimizu 1-10-20,
Yokosuka, Kanagawa, 239-8686, Japan
Nhuhai [email protected]
Department of Computer Science, National Defense Academy, Hashirimizu 1-10-20,
Yokosuka, Kanagawa, 239-8686, Japan
Hiroshi [email protected]
Department of Computer Science, National Defense Academy, Hashirimizu 1-10-20,
Yokosuka, Kanagawa, 239-8686, Japan
http://www.nda.ac.jp/cs/stuff/masaok.html
Available Online 30 June 2018.
DOI
https://doi.org/10.2991/jrnal.2018.5.1.15
Keywords
Swarm intelligence; Machine learning; Complex systems; Best-of-n problem
Abstract
The BRT agent is an algorithm that can find appropriate collective behavior
by changing the agreement contents in a trial and error manner. Computer
experiments show that it is necessary to change the agreement contents
the number of times that is proportional to the square of the number of
choices. In this paper, we attempted to shorten this search time by introducing
an agent that memorizes actions that were not able to achieve the expected
effect of what they executed. As a result, we found that search time can
be improved by just mixing a few this proposed taboo list agents.
Copyright
Copyright © 2018, the Authors. Published by ALife Robotics Corp. Ltd.
Open Access
This is an open access article under the CC BY-NC license (http://creativecommons.org/licences/by-nc/4.0/).