Title: The utility of reinforcement learning in predation of Batesian mimics

Authors: A. Tsoularis

Addresses: Business School, Leeds Metropolitan University, Bronte Hall 202, Headingley, Leeds, LS6 3QS, UK

Abstract: In this article, the focus is on modelling the predation of Batesian mimics using reinforcement learning methodology. Essentially, it is proposed that the predator be modelled as a learning automaton aided by reinforcement algorithms to select palatable prey for consumption and avoid unpalatable ones. A tentative connection between reinforcement learning and the theory of dynamic programming and optimal control is also considered. The paper concludes by offering some ideas for future research in this area.

Keywords: Batesian mimicry; reinforcement learning; learning automata; modelling; predation; mimics; dynamic programming; optimal control.

DOI: 10.1504/IJCAET.2009.028554

International Journal of Computer Aided Engineering and Technology, 2009 Vol.1 No.4, pp.494 - 501

Published online: 18 Sep 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article