In reinforcement learning, the ecosystem is usually represented like a Markov final decision process (MDP). Many reinforcements learning algorithms use dynamic programming methods.[53] Reinforcement learning algorithms don't suppose familiarity with a precise mathematical design on the MDP and they are used when specific models are infeasible. Rein