MAB Multi Armed Bandit Implementations I have implemented a naive approach, epsilon greedy, UCB and Thompson sampling I will be adding implementations for LinUCB and Contextual Thompson Sampling in the near future.