Open AccessQuantifying Learning Creativity through Simulation and Modeling of Swarm Intelligence and Neural Networks
Author(s) -
Hassan M. H. Mustafa,
Turki F. Al-Somani,
Ayoub Al-Hamadi
Publication year - 2011
Publication title -
international journal of online and biomedical engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.184
H-Index - 8
ISSN - 2626-8493
DOI - 10.3991/ijoe.v7i2.1642
Subject(s) - artificial intelligence , computer science , artificial neural network , swarm behaviour , phenomenon , similarity (geometry) , machine learning , creativity , process (computing) , swarm intelligence , foraging , field (mathematics) , particle swarm optimization , ecology , psychology , mathematics , social psychology , physics , quantum mechanics , image (mathematics) , biology , operating system , pure mathematics
This research work presents a systematic investigational study of a challenging phenomenon observed in natural world. Mainly, the study is concerned with conceptual interdisciplinary analysis and evaluation of quantified learning creativity phenomenon. In association, it deals with diverse aspects of measurable behavioral learning performance and is observed by two diverse natural biological system models (i.e. human and non-human creatures). Specifically, the studies of two biological models consider the comparison of quantified learning creativity phenomenon. The first model involves the human interactive tutoring/learning processes with environment while the other modal presents the ecological behavioral learning of swarm intelligence agents (i.e. ants) in performing the foraging process. Furthermore, a comparative study is presented which is inspired by naturally realistic models of Artificial Neural Network (ANN) and Swarm Intelligence. The obtained simulation and modeling results shows that the learning performance curves of both models behave with close similarity to each other. Precisely, the analysis and evaluation of learning performance curves of two diverse biological models revealed that both obey exponentially decayed learning curves; following least mean square (LMS) error algorithm