= Markov decision process
{wiki=Markov_decision_process}
A Markov Decision Process (MDP) is a mathematical framework used to model decision-making in situations where the outcomes are partly random and partly under the control of a decision maker. MDPs are widely used in fields like operations research, economics, robotics, and artificial intelligence, especially for reinforcement learning problems. An MDP is defined by the following components: 1. **States (S)**: A finite set of states that represent the possible situations in which an agent can find itself.
Back to article page