CS 7643 — Exam 2 High-Yield Practice Exam
CS 7643 — Georgia Tech — Exam 2 High-Yield
Exam review edition Questions with Correct
Verified Answers with Rationales
Introduction
This document provides a comprehensive Exam 2 practice and review
set for CS 7643, covering convolutional neural networks,
normalization, optimization, RNNs, LSTMs, GRUs, attention,
Transformers, generative models, and transfer learning. It contains
320 multiple-choice questions with correct answers and rationales
designed to support exam preparation and concept review.
Exam Questions and Answers
Study Instructions
• Attempt each question before viewing the correct answer.
• Focus on the mechanism behind each answer rather than
memorizing letters.
Original practice material • For study use
, CS 7643 — Exam 2 High-Yield Practice Exam
• Use the rationales to review architecture, optimization, sequence
modeling, attention, and generative-model concepts.
• Because course coverage can vary by semester, use your instructor's
syllabus and lecture materials as the final authority on what is
examinable.
High-Yield Topic Map
1. Convolution Fundamentals
2. CNN Architecture
3. CNN Receptive Fields
4. Normalization
5. Regularization
6. Optimization
7. Adam
8. Backpropagation
9. Activation Functions
10. Loss Functions
11. Data Augmentation
12. RNN Fundamentals
13. BPTT
14. Vanishing & Exploding Gradients
Original practice material • For study use
, CS 7643 — Exam 2 High-Yield Practice Exam
15. LSTM
16. GRU
17. Embeddings
18. Attention
19. Self-Attention
20. Multi-Head Attention
21. Transformers
22. Positional Encoding
23. Encoder-Decoder Attention
24. Autoregressive Language Modeling
25. Beam Search
26. Autoencoders
27. Variational Autoencoders
28. GANs
29. Diffusion Models
30. Transfer Learning
31. Evaluation & Calibration
32. Attention Complexity
33. Transformer Normalization & Residuals
Original practice material • For study use
, CS 7643 — Exam 2 High-Yield Practice Exam
34. Sequence Masking
35. Generative Modeling Objectives
Original practice material • For study use
CS 7643 — Georgia Tech — Exam 2 High-Yield
Exam review edition Questions with Correct
Verified Answers with Rationales
Introduction
This document provides a comprehensive Exam 2 practice and review
set for CS 7643, covering convolutional neural networks,
normalization, optimization, RNNs, LSTMs, GRUs, attention,
Transformers, generative models, and transfer learning. It contains
320 multiple-choice questions with correct answers and rationales
designed to support exam preparation and concept review.
Exam Questions and Answers
Study Instructions
• Attempt each question before viewing the correct answer.
• Focus on the mechanism behind each answer rather than
memorizing letters.
Original practice material • For study use
, CS 7643 — Exam 2 High-Yield Practice Exam
• Use the rationales to review architecture, optimization, sequence
modeling, attention, and generative-model concepts.
• Because course coverage can vary by semester, use your instructor's
syllabus and lecture materials as the final authority on what is
examinable.
High-Yield Topic Map
1. Convolution Fundamentals
2. CNN Architecture
3. CNN Receptive Fields
4. Normalization
5. Regularization
6. Optimization
7. Adam
8. Backpropagation
9. Activation Functions
10. Loss Functions
11. Data Augmentation
12. RNN Fundamentals
13. BPTT
14. Vanishing & Exploding Gradients
Original practice material • For study use
, CS 7643 — Exam 2 High-Yield Practice Exam
15. LSTM
16. GRU
17. Embeddings
18. Attention
19. Self-Attention
20. Multi-Head Attention
21. Transformers
22. Positional Encoding
23. Encoder-Decoder Attention
24. Autoregressive Language Modeling
25. Beam Search
26. Autoencoders
27. Variational Autoencoders
28. GANs
29. Diffusion Models
30. Transfer Learning
31. Evaluation & Calibration
32. Attention Complexity
33. Transformer Normalization & Residuals
Original practice material • For study use
, CS 7643 — Exam 2 High-Yield Practice Exam
34. Sequence Masking
35. Generative Modeling Objectives
Original practice material • For study use
