Module 3: AI and ML in Computer Engineering
1. Which of the following best captures John McCarthy’s definition of Artificial
Intelligence?
a. Designing programs to follow exact, step-by-step instructions
b. Getting a computer to do things that, when done by humans, involve intelligence
c. Creating machines that only replace physical human labor
d. Automating repetitive industrial processes without decision-making
Answer: b
2. Which feature distinguishes Artificial Intelligence from traditional programming?
a. AI executes pre-defined rules with no changes
b. AI adapts to new data and improves performance without reprogramming
c. AI is limited to solving mathematical problems only
d. AI can only run on proprietary frameworks
Answer: b
3. In supervised learning, why is labeled data essential?
a. It ensures the algorithm can be tested without human oversight
b. It directs the learning process by providing correct input-output pairs
c. It reduces storage requirements of the training dataset
d. It guarantees the model will never misclassify new inputs
Answer: b
4. Which example best illustrates unsupervised learning?
a. A model predicting exam scores using previous student records
b. A program grouping customers by purchasing habits without predefined labels
c. A weather forecast system using past weather reports
d. A chatbot trained to recognize predefined question-answer pairs
Answer: b
5. Which of the following is NOT a supervised learning algorithm?
a. Logistic regression
b. K-means clustering
c. Decision trees
d. Random forest
Answer: b
6. Why does machine learning rely heavily on iteration?
a. It allows the same dataset to be reused without errors
b. Each cycle improves predictions by adjusting based on feedback
c. It reduces the need for high-quality data
d. Iteration prevents the occurrence of overfitting
Answer: b
7. Which situation highlights the role of AI in robotics?
a. A robot following fixed assembly line instructions
b. A robot adapting its navigation path when obstacles appear
c. A robot operating only when given manual commands
d. A robot storing data in a cloud server
Answer: b
8. Which AI application is most closely tied to natural language processing
(NLP)?
a. Identifying fraudulent financial transactions
b. Understanding customer queries in a chatbot
c. Detecting tumors in medical images
d. Optimizing energy consumption in buildings
Answer: b
9. Which scenario demonstrates machine learning improving over time?
a. A calculator solving arithmetic problems correctly every time
b. A spam filter updating itself based on newly marked spam emails
c. A clock showing the same accurate time daily
d. A search engine storing cached results for faster access
Answer: b
10. Why are commercial AI frameworks attractive to companies despite their costs?
a. They allow unlimited modification of source code
b. They provide dedicated support, advanced features, and ease of use
c. They eliminate the need for skilled developers entirely
d. They guarantee faster model training in all cases
Answer: b
11. Which of the following is a limitation of unsupervised learning?
a. It cannot identify unknown patterns
b. It requires human intervention for data labeling
c. The results may be harder to interpret without clear categories
d. It always leads to overfitting
Answer: c
12. Which statement correctly distinguishes TensorFlow from PyTorch?
a. TensorFlow is primarily used in academic research, while PyTorch is more scalable
b. PyTorch is known for dynamic computation graphs, while TensorFlow emphasizes
scalability and deployment
c. Both TensorFlow and PyTorch are commercial-only frameworks
d. PyTorch is less popular due to weak community support compared to TensorFlow
Answer: b
13. Which is an advantage of Scikit-Learn compared to deep learning
frameworks?
a. It is more effective for building complex neural networks
b. It is easier to use for smaller projects and classical ML algorithms
c. It eliminates the need for preprocessing steps
d. It specializes in GPU-based large-scale training
Answer: b
14. Which example best reflects self-modifying coding in AI?
a. A code editor suggesting syntax corrections
b. A program rewriting parts of its code to improve performance after each run
c. A developer manually debugging a program
d. A computer compiling code into machine language
Answer: b
15. Which tool is NOT primarily used for deep learning tasks?
a. TensorFlow
b. Keras
c. Pandas
d. PyTorch
Answer: c
16. Which scenario best describes data mining with AI?
a. Extracting patterns from huge datasets to detect anomalies
b. Collecting raw data from surveys and interviews
c. Formatting data into rows and columns for easier access
d. Encrypting data before transferring to cloud storage
Answer: a
17. Why is feature selection important in machine learning?
a. It ensures only numerical data is used in models
b. It reduces overfitting, improves accuracy, and decreases training time
c. It adds complexity to models for more detailed results
d. It guarantees models will never make errors
Answer: b
18. Which framework is widely recognized for simplifying neural network
prototyping with user-friendly APIs?
a. Keras
b. PyTorch
c. CNTK
d. OpenNN
Answer: a
19. Which AI application demonstrates the combination of image recognition and
computer vision?
a. A self-driving car identifying traffic signs
b. A spam filter detecting phishing emails
c. A chatbot responding to customer queries
d. A cloud system retrieving stored documents
Answer: a
20. Which is a valid criticism of data visualization enhanced by AI?
a. It helps highlight outliers in datasets
b. Correlation in visuals may be mistaken as causation
c. It simplifies complex information for non-technical users
d. It allows interactive exploration of data patterns
Answer: b
