ABSTRACT
Learning is a multidisciplinary area which continuously draws attention from economists, psychologists, cognitive scientists, computer scientists, mathematicians, and neuroscientists. Among all proposed learning models, reinforcement learning seems to be the one most commonly shared by all these disciplines. Reinforcement learning was initially studied in psychology (Bush and Mosteller, 1955), then later extended to computer science by Marvin Minsky,1 games by Arthur Samuel (1901-1990), economics by John Cross (Cross, 1973, 1983), and neuroscience by Wolfram Schultz.2
