Click the cat then move the mouse around VERY SLOWLY.
Using Reinforcement learning, a type of AI, the cat will try to get to the mouse cursor. Reward: If he should ne rewarded or punished for his actions. It helps him know what to do. Speed: How fast the cat moves each calculation Epsilon: Should be from 0-1, but scratch can't do decimals in sliders. How often he should explore or exploit, or in other words, try new things or go by his feeling. Learning Rate: How fast the reward changes his calculations. If too high, he will "jump to conclusions" and not learn correctly.
