Abstract
New class of P-model absorbing ε-optimal learning automata was presented in this paper. The proposed learning automaton, Discretized Generalized Stochastic Estimator (DGSE) learning automaton, not only possesses the characteristics of the Stochastic Estimator Reward-inaction (SE RI ) learning automaton and the Discretized Generalized Pursuit Algorithm (DGPA) learning automaton, but also converges with a remarkable speed and accuracy. The asymptotic behavior of the DGSE algorithm is analyzed. Furthermore, we stick out the pitfalls in the proof of SE RI algorithm, proved the proposed DGSE algorithm to be ε-optimal, and pointed out that this proof process could be applied to prove SE RI algorithm. It’s known that the SE RI learning automaton is the fastest learning automaton up to now, whereas, the proposed DGSE learning automaton is much faster than the SE RI learning automaton. A great number of experiments and simulations verified the propose DGSE learning algorithm is quite efficient when operating in P-model stationary environment.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Papadimitriou, G.I., Sklira, M., Pomportsis, S.: A New Class of ε-optimal Learning Automata. IEEE Trans. Syst., Man, Cybern. B 34(1), 246–254 (2004)
Shutton, R.S., Barto, A.G.: Reinforcement Learning: An introduction. MIT Press, Cambridge (1998)
Tsetlin, M.L.: On the Behavior of Finite Automata in Random Media. Avtomatika i Telemekhanika 22(10), 1345–1354 (1961)
Thathachar, M.A.L., Oommen, B.J.: Automata Theory and the Modelling of Biological Systems. Academic, New York (1973)
Iraji, R., Manzuri-Shalmani, M.T., Jamalian, A.H., Beigy, H.: IJA Automaton: Experdiency and ε optimality Properties. In: Proc. 5th IEEE Int. Conf. on Congitive Informatics (ICCI 2006), pp. 617–622 (2006)
Econdomides, A.A., Kehagias, A.: The STAR automaton: Expediency and Optimality Properties. IEEE Trans. Syst., Man, Cybern. B 32(6), 723–737 (2002)
Oommen, B.J.: Absorbing and Ergodic Discretized Two-action Learning Autoamta. IEEE Trans. Syst., Man, Cybern. SMC-16(2), 282–293 (1986)
Thathachar, M.A.L., Phansalkar, V.V.: Learning the Global Maximum with Parameterized Learning Automata. IEEE Trans. Neural Networks 6, 398–406 (1995)
Beigy, H., Meybodi, M.R., Menhaj, M.B.: Adaption of Learning Rate in Backpropagation Algorithm Using Fixed Structure Learning Automata. In: Proc. of ICEE 1998, Tehran Iran, vol. 3, pp. 117–123 (1998)
Beigy, H., Meybodi, M.R.: Adaption of Momentum Factor and Steepness Parameter in Backpropagation Algorithm Using Fixed Structure Learning Automata. In: Proc. of CSICC 1999 Conference Sharif University of Technology, Tehran, Iran, pp. 117–124 (1999)
Economides, A.A.: Learning Automata Routing in Connection-oriented Networks. Int. J. Commun. Syst. 8, 225–237 (1995)
Martin, R., Tilak, O.: on ε-optimality of the Pursuit Learning Algorithm (2011)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Jiang, W. (2011). A New Class of ε-Optimal Learning Automata. In: Huang, DS., Gan, Y., Bevilacqua, V., Figueroa, J.C. (eds) Advanced Intelligent Computing. ICIC 2011. Lecture Notes in Computer Science, vol 6838. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-24728-6_16
Download citation
DOI: https://doi.org/10.1007/978-3-642-24728-6_16
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-24727-9
Online ISBN: 978-3-642-24728-6
eBook Packages: Computer ScienceComputer Science (R0)