NEW
Font size
WorksheetsUnit 1: Foundations of AI Review
Total questions: 51
Worksheet time: 26mins
Which of the following is the most accurate definition of Artificial Intelligence (AI)?
The study of how computers can mimic human movement.
The field of computer science is dedicated to creating systems that can perform tasks normally requiring human intelligence.
The science of programming robots to perform factory tasks.
The study of using databases to store information.
A vacuum cleaner robot uses its sensors to detect a wall. This action is an example of which stage in the Sense-Deliberate-Act cycle?
A. Sense
B. Deliberate
C. Act
D. Reflect
A chess-playing computer analyzes the board and calculates the optimal move to win the game. This action is an example of which stage in the Sense-Deliberate-Act cycle?
Sense
Deliberate
Act
Analyze
A factory robot lifts a car door and moves it to the assembly line. This action is an example of which stage in the Sense-Deliberate-Act cycle?
Sense
Deliberate
Act
Execute
What is the main purpose of the Turing Test?
To measure a computer's processing speed.
To determine if a machine can exhibit intelligent behavior indistinguishable from a human.
To evaluate a machine's ability to perform calculations.
To test a robot's physical dexterity.
Which of the following is an example of a system exhibiting intelligent behavior?
A. A web browser saving a user's password.
B. A word processor automatically correcting spelling errors.
C. A calculator adding a list of numbers.
D. A streaming service recommending a new movie based on your viewing history.
Which of the following best describes the difference between "knowledge" and "knowing" in an AI context?
What was a significant outcome of the early AI program, ELIZA?
It proved that machines could pass the Turing Test.
It showed that AI could have a conversation with a human.
It demonstrated that simple pattern matching could make a machine seem intelligent.
It proved that AI would never be able to understand human language.
Which of the following is a primary ethical concern related to AI systems that learn from public data?
The cost of data storage.
The time it takes to process the data.
Bias in the data that can lead to unfair or discriminatory outcomes.
The risk of the AI becoming too intelligent.
Who is considered one of the key figures in the history of AI, known for creating the Turing Test?
Albert Einstein
Alan Turing
Isaac Newton
Stephen Hawking
Which of the following is a potential societal benefit of widespread AI adoption?
Increased privacy for individuals.
Elimination of all human jobs.
Automation of dangerous or repetitive tasks.
A decrease in data security.
The Sense-Deliberate-Act cycle is a continuous loop. What happens after the "Act" stage is complete?
The process ends.
The cycle restarts with the "Sense" stage.
The system is shut down.
The system waits for human input.
What is a key risk associated with the use of AI in law enforcement?
It may make decisions too slowly.
It could perpetuate existing racial or social biases.
It might be too expensive to implement.
It may require too much human supervision.
The term "Artificial Intelligence" was coined at a Dartmouth College conference in which decade?
1950s
1960s
1970s
1980s
Which decade is often associated with the beginning of the modern computer age?
1940s
1950s
1960s
1970s
What is the main purpose of the "Think-Pair-Share" activity in a classroom setting?
To have a student work alone.
To promote collaboration and peer discussion before sharing with a larger group.
To have students individually write down their answers.
To test a student's memory of a topic.
What is the fundamental difference between machine learning and traditional programming?
ML uses a single programming language, while traditional programming uses many.
ML models learn from data without explicit instructions, while traditional programs follow pre-written rules.
ML is only used for image recognition, while traditional programming is used for all other tasks.
There is no difference; they are the same thing.
You have a dataset of historical weather information, including temperature and rainfall, to predict if it will rain tomorrow. What would be the "label" in this supervised learning scenario?
The temperature.
The amount of rainfall.
"Will it rain tomorrow?" (Yes/No).
The location of the city.
In supervised learning, the model learns from a dataset that has been:
Labeled with the correct output.
Given to the model without any labels.
Processed to remove all data.
Organized into a single category.
Which of the following is a core characteristic of a supervised learning model?
It finds patterns in data without guidance.
It learns through a system of rewards and penalties.
It uses a labeled training dataset.
It requires no data to function.
Which type of supervised learning would you use to predict the price of a stock at the end of the day?
Classification
Regression
Clustering
Association
A loan application system uses a machine learning model to decide whether to approve or deny a loan. This is an example of which type of supervised learning?
Classification
Regression
Clustering
Reinforcement Learning
You are building a model to predict the amount of electricity a house will use in a month. The output of this model will be a numerical value. What is this type of task?
Classification
Regression
Unsupervised Learning
Association
Which of the following is an example of a classification problem?
Predicting a person's age.
Forecasting the amount of sales for a company.
Identifying whether an image contains a car or a truck.
Estimating the temperature of a city.
A machine learning model is being trained to recognize handwritten digits from 0 to 9. The model's output is a category (the digit). Which type of supervised learning is this?
Regression
Classification
Association
Reinforcement Learning
The core idea of supervised learning is that the model:
Learns by trying to maximize a reward.
Learns the relationship between input data and a known output.
Learns to group data into categories without help.
Learns by following explicit instructions.
Which of the following tasks requires a labeled dataset to be used effectively?
Grouping similar news articles.
Training a supervised learning model.
Training an unsupervised learning model.
Training a reinforcement learning model.
A model predicts whether a plant is healthy or diseased. This is a task that involves predicting a:
Numerical value.
Category.
Continuous Number
Reward
Which of the following is a regression problem?
Determining if an email is spam.
Grouping similar customers.
Predicting the temperature in a room.
Identifying if a face is male or female.
What is the primary purpose of supervised learning?
To discover hidden structures in data.
To predict future values based on past data.
To learn the best course of action through trial and error.
To find associations between items in a dataset.
A model is trained to predict a student's final grade. What kind of problem is this?
A regression problem.
A classification problem.
A clustering problem.
A reinforcement learning problem.
What is the key characteristic of unsupervised learning?
It requires a labeled dataset.
It learns from a dataset with no labels.
It learns through rewards and penalties.
It predicts a numerical value.
You have a large collection of unlabeled photos and you want an AI to organize them into groups based on the people in them. Which type of unsupervised learning should you use?
Regression
Classification
Clustering
Association
A grocery store wants to analyze its sales data to find out which items are frequently purchased together, such as "bread" and "jam." This is a perfect use case for which type of unsupervised learning?
Classification
Clustering
Association
Regression
A model is given a dataset of user behavior on a website and needs to identify distinct groups of users (e.g., "heavy users," "casual users"). This is an example of:
Regression
Classification
Clustering
Reinforcement Learning
The main goal of association is to discover:
Categories within a dataset.
Numerical values.
If-then rules or patterns.
The best action to take.
An AI agent is learning to play a video game. It gets a positive score for beating a level and a negative score for failing. The "positive score" is a:
Penalty
Agent
Reward
Environment
What is the agent in a reinforcement learning scenario?
The set of rules the AI follows.
The world the AI interacts with.
The AI system that is doing the learning.
The numerical feedback the AI receives.
A company is training a robot to learn to assemble a product by rewarding it for each correct step and penalizing it for a mistake. The robot is the:
Environment
Agent
Reward
Penalty
In a reinforcement learning system, the "world" or the setting in which the agent operates is called the:
Agent
Reward
Environment
Dataset
An AI model for a self-driving car is given a reward for staying in its lane. The action of staying in the lane is part of the:
Agent
Environment
Feedback
Reward Function
What is the purpose of a Training Set in machine learning?
To evaluate the model's final performance.
To fine-tune the model's parameters.
To train the model to learn the relationships in the data.
To test the model on unseen data.
The Validation Set is primarily used to:
Test the model's final performance on new data.
Fine-tune the model to prevent overfitting.
Train the model from scratch.
Collect the initial data.
You have created a machine learning model and want to test its performance on data it has never seen before. Which dataset should you use?
Training Set
Validation Set
Test Set
Entire Dataset
Why is it important to split a dataset into training, validation, and test sets?
To make the data smaller and easier to manage.
To ensure the model can generalize to new data and isn't just memorizing the training data.
To make the training process faster.
To avoid ethical issues with the data.
A reinforcement learning agent is trying to find its way through a maze. What is its goal?
To find the fastest path through the maze.
To create a map of the maze.
To find the path that maximizes its cumulative reward.
To solve the maze without any human intervention.
Which of the following best describes the role of data in machine learning?
Data is a tool used only for traditional programming.
Data is the "fuel" that a machine learning model learns from.
Data is only used to evaluate a model's performance.
Data is not necessary for most machine learning models.
In a reinforcement learning model, a "penalty" is:
A reward for a correct action.
A negative reinforcement signal for an incorrect action.
A positive reinforcement signal for an incorrect action.
A company is using an AI to determine which customer segment a new user belongs to. The AI has to find the patterns itself, as there are no pre-existing categories. This is a use case for:
Supervised Learning
Unsupervised Learning
Reinforcement Learning
Regression
Which of the following is a classic example of a reinforcement learning problem?
An email spam filter.
A program that recommends products to a user.
A chatbot learning to have a conversation by trying to maximize user engagement.
A system that predicts the stock market.
What is the purpose of a test set?
To train the model on its main task.
To find the best parameters for the model.
To give a final, unbiased evaluation of the model's performance.
To provide the model with its initial set of data.
