Mini toolsDefinition
A function that defines the immediate feedback signal that the agent receives after taking an action in a given state.
Detailed Explanation
The reward function maps state-action pairs (or state-action-next-state transitions) to numerical rewards. It encodes the goals of the task and shapes the learning process. Designing appropriate reward functions is crucial for successful reinforcement learning, as they determine what behavior is considered optimal.
Use Cases
Training game AI, robotics tasks, recommendation systems, automated trading
