Which concept states that the future state depends only on the current state and action, not the past?

Get ready for the GARP Risk and AI Exam with flashcards and multiple choice questions. Each question comes with hints and explanations. Prepare for success!

The Markov property states that the future state depends only on the present state and the action taken, not on the past history. In an MDP, this means the transition probabilities P(s'|s,a) capture everything needed to determine the next state from the current state and action, with no dependence on how you arrived at s. This memoryless assumption lets you plan based on the current snapshot and your chosen action, rather than requiring a record of past states or actions. For example, from a given position in a grid world and a chosen move, the next position distribution is fixed regardless of the path taken to reach the current position. The other terms—policy, reward, and state—have distinct roles: policy selects actions based on the state, reward provides feedback, and state is the present description of the environment.