NEW
Font size
WorksheetsUNIT 1: Foundations of AI
Total questions: 50
Worksheet time: 25mins
A robot uses a camera to identify a door and then calculates the path to open it. Which phase of the Sense-Deliberate-Act cycle is the robot using its camera?
A. Sense
B. Deliberate
C. Act
D. Predict
Which of the following scenarios best demonstrates an AI system exhibiting intelligent behavior?
A. A simple computer program that adds a list of numbers.
B. A database application that sorts customer names alphabetically.
C. A GPS system that recalculates a driving route when it detects a road closure.
D. A video player that pauses when you click a button.
An AI model is trained on a dataset of faces for a security system. If the dataset contains a disproportionately low number of images of people with darker skin tones, which ethical concern is most likely to arise?
The system will be too slow.
The system will have a hard time connecting to the network.
The system may be bias and perform less accurately on people with darker skin tones.
The system will be too expensive to deploy.
A chat-based AI is developed to converse with users. If a human cannot tell whether they are conversing with the AI or another person, the AI has passed which famous test?
The Turing Test
The Stanford Test
The IBM Test
The Ada Lovelace Test
Which of the following describes the key contribution of the ELIZA program in early AI?
It solved complex mathematical problems.
It showed that simple pattern-matching could make a machine seem intelligent.
It proved that AI could learn autonomously.
It was the first AI to pass the Turing Test.
A smart sprinkler system uses a moisture sensor to detect that the soil is dry. It then sends a command to turn on the water. What part of this process is the moisture sensor?
Deliberate
Act
Sense
Reflect
What is the core difference between "knowledge" and "knowing" in the context of AI?
Knowledge is the raw information, while knowing is the ability to apply it.
Knowledge is the ability to apply information, while knowing is the raw information.
Knowledge is only for humans, and knowing is only for machines.
The terms are interchangeable in AI.
A company plans to use an AI to automate a task previously done by 100 employees. Which AI risk is this an example of?
Job displacement
Security breaches
Environmental impact
Algorithmic bias
Which of the following is an example of an AI benefit in a medical setting?
A. A robot vacuum cleaning a hospital floor.
B. A database storing patient records.
C. An AI system analyzing medical images to help doctors identify diseases earlier.
D. A video conferencing tool for remote patient consultations.
A self-driving car's AI decides to swerve to avoid a collision. What stage of the Sense-Deliberate-Act cycle does this decision fall under?
Sense
Deliberate
Act
Predict
Which of these characteristics is a hallmark of intelligent behavior?
The ability to calculate numbers quickly.
The ability to memorize a large amount of data.
The ability to reason, plan, and solve problems.
The ability to run on low power.
A company's AI system for hiring employees is found to be automatically rejecting applications from a certain demographic. This is a clear example of:
Efficient processing
Algorithmic bias
Job automation
Data storage issues
What was a key lesson learned from the ELIZA program?
It showed that machines could truly understand human emotion.
It highlighted the power of simple scripted responses to create the illusion of intelligence.
It proved that human conversation is too complex for AI.
It led to the development of the first autonomous robots.
Which statement is a risk associated with AI?
AI can process vast amounts of data quickly.
AI can assist in dangerous situations.
AI systems can be vulnerable to security attacks.
AI can provide personalized recommendations.
What is the primary difference between machine learning and traditional programming?
ML models are always faster.
ML models learn from data without being explicitly programmed with rules.
Traditional programs cannot be used for complex tasks.
ML is a type of traditional programming.
You are building a model to predict the price of a house. You train the model using a dataset that includes the price of each house, along with its square footage and number of bedrooms. This is an example of:
Supervised Learning
Unsupervised Learning
Reinforcement Learning
Transfer Learning
In supervised learning, what is the role of the "label" in the dataset?
It provides the input data for the model.
It provides the correct output that the model should predict.
It helps the model find hidden patterns in the data.
It provides a reward signal for correct actions.
You are developing a model to forecast the amount of rain that will fall in a city tomorrow. The model will output a numerical value in inches. This is an example of which type of supervised learning?
Classification
Regression
Clustering
Reinforcement Learning
Which of the following is a classic example of a classification problem?
Predicting the temperature tomorrow.
Grouping similar news articles together.
Identifying whether an email is "spam" or "not spam."
Forecasting a company's sales for the next quarter.
Which of the following is a classic example of a regression problem?
Identifying a handwritten digit from 0-9.
Predicting the height of a person given their age.
Grouping customers by their purchasing habits.
Deciding whether a loan application should be approved or denied.
A retail company wants to use a machine learning model to predict which customers are most likely to buy a new product. They will use historical customer data that includes a "purchased" or "did not purchase" label. Which type of model should they use?
A. A supervised learning model
B. An unsupervised learning model
C. A reinforcement learning model
D. A deep learning model
The main difference between a classification and a regression task is that a classification task predicts a category, while a regression task predicts a numerical value.
True
False
A bank uses a model to determine if a customer is likely to default on a loan. The output is "Yes" or "No". This is a supervised learning task.
True
False
A supervised learning model learns from a dataset with:
Rewards and penalties
Labeled examples
Unlabeled examples
Random exploration
A machine learning model identifies whether a picture contains a cat, a dog, or a bird. This is an example of a classification problem.
True
False
A model that predicts the number of visitors to a website per day is performing a regression task.
True
False
In supervised learning, the input data is the information given to the model, and the output data is the correct answer the model learns to predict.
True
False
Which of the following is a characteristic of unsupervised learning?
The model learns from a labeled dataset.
The model requires human supervision.
The model discovers patterns in unlabeled data.
The model learns through a system of rewards.
A streaming service wants to group its users into different segments based on their movie-watching habits, without any pre-defined categories. Which type of unsupervised learning should they use?
Regression
Classification
Clustering
Association
A retail store wants to find out which items are frequently bought together to better organize their store layout and promotions. Which type of unsupervised learning would be best for this?
A. Classification
B. Clustering
C. Association
D. Reinforcement Learning
An AI is learning to play a game. The AI receives a reward for each coin it collects. What component of the reinforcement learning process is the agent itself?
Environment
Agent
Reward
Action
In a reinforcement learning scenario, the environment is:
The AI system that is doing the learning.
The feedback signal for a correct action.
The world the agent operates in and interacts with.
A dataset of labeled examples.
A robot is being trained to navigate a complex obstacle course. It gets a penalty for bumping into obstacles. What is the goal of the robot?
To take the shortest path.
To avoid all obstacles.
To maximize its cumulative reward by avoiding penalties.
To collect all the available objects.
Which of the following is a classic example of an unsupervised learning problem?
Predicting a person's age.
Identifying fraudulent transactions.
Grouping similar news articles together.
Forecasting the temperature.
What is the purpose of the training set in a machine learning pipeline?
To test the model's final performance.
To fine-tune the model's parameters.
To allow the model to learn the relationships and patterns in the data.
To validate the model on unseen data.
You have just finished training a model. What is the primary purpose of the test set?
To allow the model to learn.
To fine-tune the model's parameters.
To provide a final, unbiased evaluation of the model's performance on new data.
To split the data into groups.
In a standard machine learning workflow, the validation set is used for:
Initial training of the model.
Final evaluation of the model's performance.
Tuning the model's hyperparameters and preventing overfitting.
Gathering the initial data.
Why is it crucial to split your data into training, validation, and test sets?
To make the training process faster.
To ensure the model can generalize to new data and isn't just memorizing the training data.
To simplify the data cleaning process.
To avoid ethical issues with the data.
A retail company uses a model to analyze shopping patterns to discover that customers who buy diapers often also buy baby wipes. This is a use case for:
Classification
Clustering
Association
Regression
The core idea of reinforcement learning is that an agent learns to make a sequence of decisions by:
Looking at a labeled dataset.
Finding patterns in an unlabeled dataset.
Receiving rewards and penalties for actions.
Following a predefined set of rules.
A machine learning model is ready for its final evaluation on unseen data. This unseen data is called the:
Training set
Validation set
Test set
Production set
In a reinforcement learning environment, the agent's goal is to:
Perform actions that maximize its final reward.
Memorize the environment's layout.
Find patterns in the data it is given.
Complete a task as fast as possible.
A model that finds groups of similar-looking galaxies in astronomy images is using:
Unsupervised Learning
Supervised Learning
Reinforcement Learning
Regression
Which type of machine learning requires a system of rewards and penalties?
Supervised Learning
Unsupervised Learning
Reinforcement Learning
Both A and C
What is the purpose of a validation set?
To check the model's performance on the final product.
To ensure the model doesn't overfit to the training data.
To determine the size of the training set.
To provide the model with its initial instructions.
You have a dataset that includes historical weather data, park attendance numbers, and the corresponding wait times for the "Galactic Coaster" ride. Predicting the wait time for the ride in minutes is an example of which type of machine learning problem, and what type of model would you use?
This is a supervised learning problem, and a regression model would be used.
This is an unsupervised learning problem, and a clustering model would be used.
This is a supervised learning problem, and a classification model would be used.
This is an unsupervised learning problem, and a regression model would be used.
Supervised Learning is:
A type of machine learning where the model is trained on labeled data.
A type of learning where the model learns without any data.
A type of machine learning that does not require any supervision.
A type of learning that only uses unsupervised data.
Labeled data in the context of machine learning refers to:
Data that has been tagged with the correct output or category.
Data that is missing values.
Data that is collected from sensors.
Data that is encrypted for security purposes.
Regression in machine learning is:
A technique used to predict continuous values.
A method for classifying images.
A process for clustering data points.
A way to sort data alphabetically.
In machine learning, what type of values does Regression predict?
Numbers (continuous values).
Categories (discrete labels).
Images.
Text.
