Markov Decision Process (MDP): A mathematical framework used to model decision-making in environments where the next state depends only on the current state and the action taken, used in reinforcement learning.