
For FullText PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.

A Constructive Compound Neural Networks. II Application to Artificial Life in a Competitive Environment
Jianjun YAN Naoyuki TOKUDA Juichi MIYAMICHI
Publication
IEICE TRANSACTIONS on Information and Systems
Vol.E83D
No.4
pp.845856 Publication Date: 2000/04/25 Online ISSN:
DOI: Print ISSN: 09168532 Type of Manuscript: PAPER Category: Artificial Intelligence, Cognitive Science Keyword: neural networks construction, artificial life, fuzzy logic, genetic algorithm, reinforcement learning,
Full Text: PDF>>
Summary:
We have developed a new efficient neural networkbased algorithm for Alife application in a competitive world whereby the effects of interactions among organisms are evaluated in a weak form by exploiting the position of nearest food elements into consideration but not the positions of the other competing organisms. Two online learning algorithms, an instructive ASL (adaptive supervised learning) and an evaluative feedbackoriented RL (reinforcement learning) algorithm developed have been tested in simulating Alife environments with various neural network algorithms. The constructive compound neural network algorithm FuzGa guided by the ASL learning algorithm has proved to be most efficient among the methods experimented including the classical constructive cascaded CasCor algorithm of [18],[19] and the fixed nonconstructive fuzzy neural networks. Adopting an adaptively selected best sequence of feedback action period Δα which we have found to be a decisive parameter in improving the network efficiency, the ASLguided FuzGa had a performance of an averaged fitness value of 541.8 (standard deviation 48.8) as compared with 500(53.8) for ASLguided CasCor and 489.2 (39.7) for RLguided FuzGa. Our FuzGa algorithm has also outperformed the CasCor in time complexity by 31.1%. We have elucidated how the dimensionless parameter food availability F_{A} representing the intensity of interactions among the organisms relates to a best sequence of the feedback action period Δα and an optimal number of hidden neurons for the given configuration of the networks. We confirm that the present solution successfully evaluates the effect of interactions at a larger F_{A}, reducing to an isolated solution at a lower value of F_{A}. The simulation is carried out by thread functions of Java by ensuring the randomness of individual activities.


