A Bayesian Learning Automaton for Solving Two-Armed Bernoulli Bandit Problems | IEEE Conference Publication | IEEE Xplore