[STUDY GUIDE] • HIGH-YIELD PRACTICE & REVIEW EDITION
CS7643 Quiz 5 Questions & Answer Key
Solutions Latest 2026/2027 Edition
Comprehensive Examination Question Bank • Concept Mapping
TOTAL QUESTIONS EXAM TOPICS ANSWERS
22 Questions 6 Modules 100% Verified
DOCUMENT OVERVIEW
This document contains 22 verified questions with answers regarding advanced topics in machine learning, specifically
focusing on encoder-decoder architectures, attention mechanisms, and neural networks. Each question is paired with a correct
answer, providing students with a comprehensive study resource suitable for exam preparation, coursework review, and
deepening their understanding of modern natural language processing techniques.
EXAM BLUEPRINT & TOPIC DISTRIBUTION
Systematic breakdown of subject domains and exam coverage.
Topic Module Scope & Core Focus Questions Share (%)
Softmax Function and Attention Explores the properties and applications of the softmax function in
Mechanisms attention mechanisms. 7 Qs 31.8%
Covers the encoder-decoder architecture and its application in
Sequence-to-Sequence Models sequence-to-sequence prediction tasks. 4 Qs 18.2%
Examines different types of attention functions, including hard and
Attention Functions soft attention. 5 Qs 22.7%
Focuses on the beam search algorithm and its role in optimizing
Beam Search Algorithm sequence generation. 2 Qs 9.1%
Discusses the foundational principles of transformer models in
Transformer Architecture language processing. 2 Qs 9.1%
Word Embeddings and Positional Investigates the concepts of word embeddings and their positional
Encoding encodings in neural networks. 2 Qs 9.1%
Total Exam Coverage 6 Integrated Topic Modules 22 Qs 100.0%
Confidential • Student Study Edition • Practice & Review Guide Page 1 of 8
, STUDENT STUDY & MASTERY EDITION PRACTICE & REVIEW GUIDE
TOPIC 1: SOFTMAX FUNCTION AND ATTENTION MECHANISMS
7 Questions • 31.8% of Exam • Explores the properties and applications of the softmax function in attention mechanisms.
QUESTION 1
Properties of the softmax function
Answer: 1.) Softmax is permutation equivariant (permutation of input leads to the same permutation of output) 2.)
Softmax is not linear 3.) Softmax is differentiable 4.) Softmax interpolates between a distribution that selects index
uniformly at random and dist that selects the argmax index with probability 1.
QUESTION 2
What does softmax allow?
Answer: Allows random sampling of numbers from original set with a distribution that depends on the values from the
numbers of the original set.
QUESTION 3
Softmax Attention
Answer: - Given a set. of vectors "U" and a query vector "Q" - We can select the most similar vector to "Q" via
p=softmax(U * Q)
QUESTION 4
What does Softmax attention do?
Answer: Gives a method of differentially selecting a vector from a set.
QUESTION 5
Forms of attention
Answer: Hard and soft.
Confidential • Student Study Edition • Practice & Review Guide Page 2 of 8
CS7643 Quiz 5 Questions & Answer Key
Solutions Latest 2026/2027 Edition
Comprehensive Examination Question Bank • Concept Mapping
TOTAL QUESTIONS EXAM TOPICS ANSWERS
22 Questions 6 Modules 100% Verified
DOCUMENT OVERVIEW
This document contains 22 verified questions with answers regarding advanced topics in machine learning, specifically
focusing on encoder-decoder architectures, attention mechanisms, and neural networks. Each question is paired with a correct
answer, providing students with a comprehensive study resource suitable for exam preparation, coursework review, and
deepening their understanding of modern natural language processing techniques.
EXAM BLUEPRINT & TOPIC DISTRIBUTION
Systematic breakdown of subject domains and exam coverage.
Topic Module Scope & Core Focus Questions Share (%)
Softmax Function and Attention Explores the properties and applications of the softmax function in
Mechanisms attention mechanisms. 7 Qs 31.8%
Covers the encoder-decoder architecture and its application in
Sequence-to-Sequence Models sequence-to-sequence prediction tasks. 4 Qs 18.2%
Examines different types of attention functions, including hard and
Attention Functions soft attention. 5 Qs 22.7%
Focuses on the beam search algorithm and its role in optimizing
Beam Search Algorithm sequence generation. 2 Qs 9.1%
Discusses the foundational principles of transformer models in
Transformer Architecture language processing. 2 Qs 9.1%
Word Embeddings and Positional Investigates the concepts of word embeddings and their positional
Encoding encodings in neural networks. 2 Qs 9.1%
Total Exam Coverage 6 Integrated Topic Modules 22 Qs 100.0%
Confidential • Student Study Edition • Practice & Review Guide Page 1 of 8
, STUDENT STUDY & MASTERY EDITION PRACTICE & REVIEW GUIDE
TOPIC 1: SOFTMAX FUNCTION AND ATTENTION MECHANISMS
7 Questions • 31.8% of Exam • Explores the properties and applications of the softmax function in attention mechanisms.
QUESTION 1
Properties of the softmax function
Answer: 1.) Softmax is permutation equivariant (permutation of input leads to the same permutation of output) 2.)
Softmax is not linear 3.) Softmax is differentiable 4.) Softmax interpolates between a distribution that selects index
uniformly at random and dist that selects the argmax index with probability 1.
QUESTION 2
What does softmax allow?
Answer: Allows random sampling of numbers from original set with a distribution that depends on the values from the
numbers of the original set.
QUESTION 3
Softmax Attention
Answer: - Given a set. of vectors "U" and a query vector "Q" - We can select the most similar vector to "Q" via
p=softmax(U * Q)
QUESTION 4
What does Softmax attention do?
Answer: Gives a method of differentially selecting a vector from a set.
QUESTION 5
Forms of attention
Answer: Hard and soft.
Confidential • Student Study Edition • Practice & Review Guide Page 2 of 8
