AWS Certified AI Practitioner practice questions
Anyone who works with AI and ML on AWS, technical or not
What the AIF-C01 exam covers
Current exam guideSample AIF-C01 questions
Five of the 25 practice questions, each with its answer and explanation.
Fundamentals of AI and ML
A company wants to predict house prices from past sales where each sale has a known price. Which type of machine learning fits?
- Unsupervised learning
- Image generation
- Supervised learning
- Reinforcement learning
Show the answer
C. Supervised learning
The past sales are labelled with the answer (the price), which is what supervised learning needs. Predicting a number is a regression task.
Fundamentals of AI and ML
A retailer wants to group customers into segments, but has no predefined segment labels. Which approach fits?
- Reinforcement learning
- Unsupervised clustering
- Translation
- Supervised classification
Show the answer
B. Unsupervised clustering
Clustering finds natural groupings in data without labels.
Fundamentals of AI and ML
What is it called when a trained model is used to make predictions on new data?
- Feature engineering
- Inference
- Training
- Labelling
Show the answer
B. Inference
Inference is using a trained model to produce outputs. Training is how the model is built.
Fundamentals of AI and ML
Which AWS service is used to build, train and deploy custom machine learning models?
- AWS Glue
- Amazon QuickSight
- Amazon Rekognition
- Amazon SageMaker AI
Show the answer
D. Amazon SageMaker AI
SageMaker AI covers the full lifecycle of custom models. Rekognition is a ready-made image analysis service.
Fundamentals of AI and ML
A model scores very well on its training data but poorly on new data. What is this?
- Underfitting
- Data drift
- Low latency
- Overfitting
Show the answer
D. Overfitting
Overfitting means the model has memorised the training data instead of learning patterns that generalise.