reinforcement-learning

Definition

Credit Assignment Problem

In reinforcement learning, the credit assignment problems describes the issue of assigning rewards/credits to singular steps in a trajectory.

Example: Which move in a Chess game was the brilliant one that won the game?