UGC NET Dec.2023 | CS - Artificial Intelligence | Top 10 Question analysis! Must watch!!

Introduction

Welcome to the session dedicated to all UGC NET aspirants at Professor Academy! In today’s discussion, we will explore the Top 10 AI Questions that are crucial for your examination preparation. These questions have been repeated in previous years of the UGC NET examination, thus familiarizing yourself with these will cover approximately 80% of the Artificial Intelligence syllabus. Let’s dive in!

Question 1:

The A algorithm is optimal when:*

• Option 1: It always finds the solution with the lowest total cost if the heuristic h(n) is admissible.
• Option 2: It always finds the solution with the highest total cost if the heuristic h(n) is admissible.
• Option 3: It finds the solution with the total cost if the heuristic h(n) is not admissible.
• Option 4: It always finds the solution with the highest total cost if the heuristic h(n) is not admissible.

Answer: Option 1 is correct. The A* algorithm is a popular pathfinding algorithm that efficiently searches for the shortest path between two points on a graph. Its optimality is guaranteed when the heuristic function is admissible, meaning it does not overestimate the cost to reach the goal node.

Question 2:

Which artificial intelligence technique enables computers to understand the associations and relationships between objects and events?

• Option 1: Heuristic processing
• Option 2: Cognitive science
• Option 3: Relational symbolism
• Option 4: Pattern matching

Answer: Option 4, Pattern Matching, is correct. This technique identifies similarities or patterns within data and is used in various AI applications to recognize relationships between objects or events.

Question 3:

Where does the value of Alpha Beta search get updated?

• Option 1: Along the path of search
• Option 2: Initial state itself
• Option 3: At the end
• Option 4: None of these

Answer: Option 1 is the correct answer. The values of Alpha and Beta in Alpha Beta pruning get updated along the path of the search, optimizing the Minimax algorithm by reducing the number of nodes evaluated.

Question 4:

Calculate the total output for the input values given weights: 1, 2, 3, 4 and inputs: 5, 7, 10, 30 with a constant of proportionality equal to 3.

Answer: The output is 57. You multiply the weights by the inputs and then apply the transfer function.

Question 5:

Which AI system mimics the evolutionary process?

• Option A: Self-organizing neural network
• Option B: Back propagation neural network
• Option C: Genetic algorithm
• Option D: Forward propagation neural network

Answer: Option C; Genetic Algorithm is correct. This AI system uses mechanisms such as selection, crossover, and mutation to find increasingly better solutions through a natural selection process.

Question 6:

Identify the learning algorithms that represent classification.

• Option A: Logistic regression
• Option B: Back propagation
• Option C: Linear regression
• Option D: Forward propagation

Answer: Correct options are A and B only: Logistic regression and Back propagation are classification algorithms.

Question 7:

Match the following:

• 1. Text planning
• 2. Sentence planning
• 3. Sentence generation
• 4. Map the input to useful representations

Answer: 1 - NLG; 2 - NLG; 3 - NLG; 4 - NLU

Question 8:

What are the characteristics of artificial neural networks?

• Option A: It is essentially a machine learning algorithm.
• Option B: It is useful for large data sets.
• Option C: They extract features with no input from the program.
• Option D: These are modeled on the human brain.

Answer: All statements are correct.

Question 9:

Overfitting is expected when:

• Option 1: Training and test set errors decrease.
• Option 2: Training set error decreases but test set error increases.
• Option 3: Both errors increase with training.
• Option 4: Both errors remain constant.

Answer: Option 2 is correct. Overfitting occurs when the training error decreases with training iterations, but the test set error increases.

Question 10:

Which agent deals with happy and unhappy states?

• Option 1: Goal-based agent
• Option 2: Learning agent
• Option 3: Model-based agent
• Option 4: Utility-based agent

Answer: Option 4, Utility-based agent, is correct. This agent maximizes satisfaction or utility in its decisions.

Keywords

• A* algorithm
• Pathfinding
• Alpha Beta search
• Genetic algorithm
• Classification algorithms
• Overfitting
• Utility-based agent
• Neural networks

FAQ

1. What is the A algorithm used for?*

   • The A* algorithm is a popular pathfinding and graph traversal algorithm that finds the shortest path between two nodes.

2. What does pattern matching in AI do?

   • Pattern matching analyzes patterns or similarity in data to recognize associations between objects and events.

3. How does Alpha Beta pruning work?

   • Alpha Beta pruning is an optimization technique for the minimax algorithm that reduces the number of nodes evaluated in search trees.

4. What is overfitting in machine learning?

   • Overfitting occurs when a model learns the training data too well, including noise, resulting in poor generalization on unseen data.

5. What is a utility-based agent?

   • A utility-based agent is an AI system that evaluates different states based on their utility or desirability, making decisions to maximize satisfaction.