Crash Course-AI#9

What is the main idea behind reinforcement learning?

• Learning from a teacher's answers

• Copying human behavior

• Finding patterns in data

• Learning by trial-and-error

In reinforcement learning, what is an 'agent' responsible for?

• Providing rewards

• Making predictions or performing actions

• Assigning values

• Setting the environment

What is the 'state' in reinforcement learning?

• The final reward

• The current inputs to the agent

• The actions taken by the agent

• The environment setup

Why is the trade-off between exploration and exploitation important in reinforcement learning?

• To ensure the agent always takes the same path

• To make the environment more complex

• To avoid giving any rewards

• To balance between trying new actions and using known paths

What can make reinforcement learning problems more complicated?

• Avoiding the use of value functions

• Introducing different types of rewards and changing environments

• Having a static environment

• Using only one type of reward