A study of various n-armed bandit algorithms. Based on Sutton and Barto, Reinforcement Learning, An Introduction.
We compare various algorithms to solve the n-armed bandit problem.
In industry, we write code to make cool projects to sell to people. In academia, we write code to make cool graphs. If you want to look at the graphs without running this code yourself (it does take some time!), take a look at the published http://kuitang.github.com/banditology/