AWS Certified AI Practitioner (AIF-C01)
1. What is Artificial Intelligence (AI)?
A. A programming language used for data analysis
B. The simulation of human intelligence processes by machines
C. A database management system
D. A cloud computing service
✓ Answer: B AI refers to the simulation of human intelligence in machines that
are programmed to think and learn like humans.
2. Which of the following best describes Machine Learning (ML)?
A. Writing explicit rules for a computer to follow
B. A subset of AI that enables systems to learn from data without being
explicitly programmed
C. A type of database query language
D. A network security protocol
✓ Answer: B Machine Learning is a subset of AI that enables systems to
automatically learn and improve from experience without being explicitly
programmed.
3. What is Deep Learning?
A. A subset of ML using neural networks with many layers
B. A type of relational database
C. A cloud storage solution
D. A data encryption method
✓ Answer: A Deep Learning is a subset of Machine Learning that uses artificial
neural networks with multiple layers (hence 'deep') to model complex patterns.
4. Which type of ML involves learning from labeled data?
A. Unsupervised learning
B. Reinforcement learning
, C. Supervised learning
D. Semi-supervised learning
✓ Answer: C Supervised learning uses labeled training data where the desired
output is known to train models.
5. Which type of ML involves learning from unlabeled data to find patterns?
A. Supervised learning
B. Unsupervised learning
C. Reinforcement learning
D. Transfer learning
✓ Answer: B Unsupervised learning finds hidden patterns or intrinsic structures
in unlabeled data.
6. In reinforcement learning, what does an agent maximize?
A. Accuracy
B. Data volume
C. Cumulative reward
D. Model size
✓ Answer: C In reinforcement learning, an agent learns by interacting with an
environment to maximize cumulative reward.
7. What is a neural network?
A. A computer network protocol
B. A computing system inspired by the human brain's structure
C. A type of relational database
D. A cloud VPN service
✓ Answer: B A neural network is a series of algorithms that attempts to
recognize underlying relationships in data through a process that mimics the way
the human brain operates.
8. What is Natural Language Processing (NLP)?
A. Processing numerical data
B. AI technology that enables computers to understand and generate
human language
C. A network communication protocol
D. A data compression technique
✓ Answer: B NLP is a branch of AI focused on the interaction between
computers and humans through natural language.
9. What is Computer Vision?
A. A hardware component for computers
, B. An AI field that trains computers to interpret visual information from
the world
C. A software IDE for developers
D. A cloud storage service
✓ Answer: B Computer Vision is a field of AI that enables computers to identify
and understand objects and people in images and videos.
10. What is Transfer Learning?
A. Moving data between AWS regions
B. Reusing a pre-trained model on a new but related problem
C. Transferring model weights to production
D. Copying datasets between S3 buckets
✓ Answer: B Transfer learning is a technique where a model trained on one
task is repurposed as the starting point for a model on a different but related task.
11. What does the term 'training data' refer to in ML?
A. Documentation for ML engineers
B. The dataset used to fit a machine learning model
C. Test cases for software
D. GPU specifications
✓ Answer: B Training data is the dataset used to train a machine learning
model to make predictions or decisions.
12. What is overfitting in machine learning?
A. When a model performs well on training data but poorly on new data
B. When a model is too simple to capture patterns
C. When there is too much training data
D. When the model runs too slowly
✓ Answer: A Overfitting occurs when a model learns the training data too well,
including noise, and fails to generalize to new unseen data.
13. What is underfitting in machine learning?
A. A model that has too many parameters
B. A model that is too complex
C. A model that is too simple to capture underlying patterns in the data
D. A model trained on too much data
✓ Answer: C Underfitting occurs when a model is too simple to capture the
underlying patterns in the training data, resulting in poor performance.
14. What is a hyperparameter in ML?
A. A parameter learned from training data
, B. A configuration setting set before training that controls the training
process
C. An output prediction
D. A data preprocessing step
✓ Answer: B Hyperparameters are parameters whose values are set before the
learning process begins, as opposed to model parameters which are derived
from training.
15. What is the purpose of a validation dataset?
A. To train the model
B. To tune hyperparameters and evaluate model performance during
training
C. To deploy the model
D. To store raw data
✓ Answer: B A validation dataset is used to provide an unbiased evaluation
during model training to tune hyperparameters and prevent overfitting.
16. What is cross-validation?
A. Validating a model on multiple cloud providers
B. A technique to assess model performance by training and testing on
different subsets of data
C. Comparing two different ML algorithms
D. A data encryption technique
✓ Answer: B Cross-validation is a statistical technique that involves partitioning
data into subsets to evaluate model performance and generalizability.
17. What is a confusion matrix?
A. A visualization of training data distribution
B. A table used to evaluate classification model performance showing
true vs predicted values
C. A matrix for storing model weights
D. A type of neural network layer
✓ Answer: B A confusion matrix is a table used to evaluate the performance of
a classification model by comparing actual and predicted classifications.
18. What does precision measure in ML classification?
A. The percentage of actual positives correctly identified
B. Of all predicted positives, the percentage that are actually positive
C. The overall accuracy of the model
D. The speed of the model
1. What is Artificial Intelligence (AI)?
A. A programming language used for data analysis
B. The simulation of human intelligence processes by machines
C. A database management system
D. A cloud computing service
✓ Answer: B AI refers to the simulation of human intelligence in machines that
are programmed to think and learn like humans.
2. Which of the following best describes Machine Learning (ML)?
A. Writing explicit rules for a computer to follow
B. A subset of AI that enables systems to learn from data without being
explicitly programmed
C. A type of database query language
D. A network security protocol
✓ Answer: B Machine Learning is a subset of AI that enables systems to
automatically learn and improve from experience without being explicitly
programmed.
3. What is Deep Learning?
A. A subset of ML using neural networks with many layers
B. A type of relational database
C. A cloud storage solution
D. A data encryption method
✓ Answer: A Deep Learning is a subset of Machine Learning that uses artificial
neural networks with multiple layers (hence 'deep') to model complex patterns.
4. Which type of ML involves learning from labeled data?
A. Unsupervised learning
B. Reinforcement learning
, C. Supervised learning
D. Semi-supervised learning
✓ Answer: C Supervised learning uses labeled training data where the desired
output is known to train models.
5. Which type of ML involves learning from unlabeled data to find patterns?
A. Supervised learning
B. Unsupervised learning
C. Reinforcement learning
D. Transfer learning
✓ Answer: B Unsupervised learning finds hidden patterns or intrinsic structures
in unlabeled data.
6. In reinforcement learning, what does an agent maximize?
A. Accuracy
B. Data volume
C. Cumulative reward
D. Model size
✓ Answer: C In reinforcement learning, an agent learns by interacting with an
environment to maximize cumulative reward.
7. What is a neural network?
A. A computer network protocol
B. A computing system inspired by the human brain's structure
C. A type of relational database
D. A cloud VPN service
✓ Answer: B A neural network is a series of algorithms that attempts to
recognize underlying relationships in data through a process that mimics the way
the human brain operates.
8. What is Natural Language Processing (NLP)?
A. Processing numerical data
B. AI technology that enables computers to understand and generate
human language
C. A network communication protocol
D. A data compression technique
✓ Answer: B NLP is a branch of AI focused on the interaction between
computers and humans through natural language.
9. What is Computer Vision?
A. A hardware component for computers
, B. An AI field that trains computers to interpret visual information from
the world
C. A software IDE for developers
D. A cloud storage service
✓ Answer: B Computer Vision is a field of AI that enables computers to identify
and understand objects and people in images and videos.
10. What is Transfer Learning?
A. Moving data between AWS regions
B. Reusing a pre-trained model on a new but related problem
C. Transferring model weights to production
D. Copying datasets between S3 buckets
✓ Answer: B Transfer learning is a technique where a model trained on one
task is repurposed as the starting point for a model on a different but related task.
11. What does the term 'training data' refer to in ML?
A. Documentation for ML engineers
B. The dataset used to fit a machine learning model
C. Test cases for software
D. GPU specifications
✓ Answer: B Training data is the dataset used to train a machine learning
model to make predictions or decisions.
12. What is overfitting in machine learning?
A. When a model performs well on training data but poorly on new data
B. When a model is too simple to capture patterns
C. When there is too much training data
D. When the model runs too slowly
✓ Answer: A Overfitting occurs when a model learns the training data too well,
including noise, and fails to generalize to new unseen data.
13. What is underfitting in machine learning?
A. A model that has too many parameters
B. A model that is too complex
C. A model that is too simple to capture underlying patterns in the data
D. A model trained on too much data
✓ Answer: C Underfitting occurs when a model is too simple to capture the
underlying patterns in the training data, resulting in poor performance.
14. What is a hyperparameter in ML?
A. A parameter learned from training data
, B. A configuration setting set before training that controls the training
process
C. An output prediction
D. A data preprocessing step
✓ Answer: B Hyperparameters are parameters whose values are set before the
learning process begins, as opposed to model parameters which are derived
from training.
15. What is the purpose of a validation dataset?
A. To train the model
B. To tune hyperparameters and evaluate model performance during
training
C. To deploy the model
D. To store raw data
✓ Answer: B A validation dataset is used to provide an unbiased evaluation
during model training to tune hyperparameters and prevent overfitting.
16. What is cross-validation?
A. Validating a model on multiple cloud providers
B. A technique to assess model performance by training and testing on
different subsets of data
C. Comparing two different ML algorithms
D. A data encryption technique
✓ Answer: B Cross-validation is a statistical technique that involves partitioning
data into subsets to evaluate model performance and generalizability.
17. What is a confusion matrix?
A. A visualization of training data distribution
B. A table used to evaluate classification model performance showing
true vs predicted values
C. A matrix for storing model weights
D. A type of neural network layer
✓ Answer: B A confusion matrix is a table used to evaluate the performance of
a classification model by comparing actual and predicted classifications.
18. What does precision measure in ML classification?
A. The percentage of actual positives correctly identified
B. Of all predicted positives, the percentage that are actually positive
C. The overall accuracy of the model
D. The speed of the model
