Home // COGNITIVE 2015, The Seventh International Conference on Advanced Cognitive Technologies and Applications // View article

Generating Opinion Agent-based Models by Structural Optimisation

Authors:
Alwyn Husselmann

Keywords: Agent-based models; Optimisation; Domain-specific languages; Multi-stage Programming

Abstract:
Agent-based modelling has enjoyed a significant increase in research effort in recent years. Particular efforts in the combination of it with optimisation algorithms have allowed the automatic generation of interesting system-level behaviors. The vast majority of these efforts have focussed on parametric optimisation, whereby the structure of a model remains user-defined, and parameters are systematically calibrated. Fairly little effort has been expended in investigations towards combinatorial optimisation in the context of agent-based modelling. The author has previously shown that it is possible to combine the use of a domain-specific language (DSL) and the multi- stage programming paradigm to provide a platform suitable for extension using an evolutionary algorithm. This combination was important to allow run-time code generation, while using just-in-time compiling. The entire process is extremely compute intensive, but can be successfully mitigated using such a language. In this article, three experiments are carried out using this language, and the efficacy of this optimisation is discussed.

Pages: 108 to 114

Copyright: Copyright (c) IARIA, 2015

Publication date: March 22, 2015

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-390-2

Location: Nice, France

Dates: from March 22, 2015 to March 27, 2015