سال انتشار: ۱۳۸۳

محل انتشار: دهمین کنفرانس سالانه انجمن کامپیوتر ایران

تعداد صفحات: ۶

نویسنده(ها):

hamid beigy – computer engineering department sharif university of technology
m.r meybodi – soft computing laboratory computer engineering department amirkabir university of tech

چکیده:

in this paper we intorduce a new state dependent nonstationary environment and study the asymptotic behavior of SLrI learning algorithm operating under the proposed environment it is shown that the SLR-I automaton operating in the proposed nontationary environment equalizes the expected penalty strengths of actions this model was motivated by applications of learning automata in call admission in cellular networks