Knowledge quiz

It's time to put what you've learned to the test, get 8 right to pass this week.

In the banana hunting game, why does the monkey jump even if there is no barrel nearby?

Choose the correct answer.

Unnecessary jumping does not lead to any punishment.

The monkey has a very high learning rate.

Not jumping was punished.

The agent alternates between the jumping and not jumping actions.

What is the second step in reinforcement learning?

Choose the correct answer.

Try action

Capture state

Receive reward (or punishment)

Adjust strategy

Which principle is essential to reinforcement learning?

Choose the correct answer.

The agent receives a reward or punishment at certain points in time and thus learns to assess what value an action has in a certain state.

The agent receives instructions on how to behave in a particular state.

A programmer has specified the best action for the agent in a given state. If it follows their recommendation, it is rewarded.

Reward and punishment are balanced in their frequency of occurrence.

What influence does a reward typically have on the learning process in reinforcement learning?

Choose the correct answer.

All actions that contributed to the reward are shown more frequently in the future.

Only the action that directly led to the reward will be shown more often in the future.

None, as long as the agent has not also experienced a punishment.

All actions that did not contribute to obtaining the reward will be shown less frequently in the future.

Which of the following applications is considered an example of AI?

There are two correct answers.

Style change filters for a photo, which for example transfer the art style of a well-known painter.

Music recommendations based on one's musical taste.

A huge database that contains many millions of videos and enables their playback.

Calculating the mean value of a series of numbers.

Which pseudocode can be used to describe the reward and punishment for the agent in the banana hunt game?

Choose the correct answer.

if touching barrel then -10 else 1

if touching mouse pointer then -10 else 1

wait until touching barrel

if -10 > 1 then touching barrel else touching floor

Which of the following statements about the model in the Banana Hunt example are true?

There are three correct answers.

The values for a new state are initialized in the model whenever this state occurs for the first time.

A positive value for a specific action in a specific state in the model means that the monkey has received more rewards than punishments for this action in this state.

The model is updated after each action.

The model table consists of 2 columns, the values for the "jump" and the "do nothing" action.

The model is initialized at the beginning of the program and updated once after the program is stopped.

What is the name of a system that can help people solve more complex problems by deriving recommended actions from a knowledge base?

Choose the correct answer.

Expert system

Experiment system

Agent

Deep Blue machine

What is weak AI?

Choose the correct answer.

Software that can solve a specific problem

Software that solves problems slower than humans

Computers with artificial intelligence and very high power consumption

Software that does not require training data to solve specific problems

10.

What is the relationship between machine learning and artificial intelligence?

Choose the correct answer.

Machine learning is a subfield of artificial intelligence.

Artificial intelligence is a subfield of machine learning.

Machine learning is another term for artificial intelligence.

Machine learning describes the opposite of artificial intelligence.