Artificial Intelligence - Important Questions with Detailed Explanations

Module 1: Introduction to Artificial Intelligence (4 Hrs)

Q1. Define Artificial Intelligence.

●​ Artificial Intelligence (AI) is a branch of computer science focused on creating machines

that can simulate human intelligence. This includes tasks like learning from experience
(machine learning), understanding natural language, recognizing patterns, and making
decisions.​

Q2. Discuss the history of AI.

●​ 1950s: Alan Turing developed the concept of the Turing Test to measure machine
intelligence.​

●​ 1956: AI as a field was officially born at the Dartmouth Conference.​

●​ 1960s–70s: Rule-based systems and symbolic reasoning emerged.​

●​ 1980s: Expert Systems became popular, e.g., MYCIN for medical diagnosis.​

●​ 1997: IBM's Deep Blue defeated world chess champion Garry Kasparov.​

●​ 2010s–Present: Rise of deep learning, big data, and AI in everyday applications (voice
assistants, recommendation systems).​

Q3. What are the components of an AI program?

●​ Knowledge Base: Contains facts and rules about the domain.​

●​ Inference Engine: Applies logical rules to the knowledge base to deduce new
information.​

●​ Learning Component: Allows the system to improve over time based on experience.​

●​ User Interface: Facilitates communication between user and system.​

Q4. Explain the categorization of intelligent systems.

 ●​ Simple Reflex Agents: Make decisions based only on current percepts (e.g.,
thermostat).​

●​ Model-based Reflex Agents: Maintain internal state to deal with partial observability.​

●​ Goal-based Agents: Make decisions based on a desired goal (e.g., route planning).​

●​ Utility-based Agents: Aim to maximize a utility function (e.g., self-driving cars

optimizing comfort and safety).​

●​ Learning Agents: Can learn from interactions and improve performance (e.g.,
recommendation engines).​

Q5. What are the sub-areas of AI?

●​ Machine Learning: Algorithms that learn patterns from data.​

●​ Natural Language Processing: Enables machines to understand human language.​

●​ Computer Vision: Allows systems to interpret visual input.​

●​ Robotics: Combines AI with mechanical systems.​

●​ Expert Systems: Rule-based systems for specialized tasks.​

Q6. Applications of AI:

●​ Healthcare: AI helps in diagnosis, drug discovery, and robotic surgery.​

●​ Retail: Personalized recommendations and inventory management.​

●​ Finance: Fraud detection and automated trading.​

●​ Transportation: Self-driving cars and route optimization.​

Q7. Current Trends in AI:

●​ Generative AI (e.g., ChatGPT, DALL-E): Capable of creating human-like text/images.​

●​ Explainable AI: Focus on transparency in decision-making.​

 ●​ AI Ethics: Ensures fairness, accountability, and bias-free systems.​

Module 2: Intelligent Agents (4 Hrs)

Q1. What are Agents and Environments?

●​ Agent: An entity that perceives its environment through sensors and acts upon it using
actuators.​

●​ Environment: The external context in which an agent operates.​

Example: A robot vacuum (agent) senses dirt (environment) and cleans it.

Q2. Define Rationality in AI.

●​ A rational agent chooses actions that maximize expected performance measure based
on its knowledge.​

●​ Rationality ≠ perfection; it depends on the agent's percept sequence, knowledge, and

computational resources.​

Q3. Describe the nature of environments.

●​ Fully vs Partially Observable: Can the agent access the complete state?​

●​ Deterministic vs Stochastic: Is the next state predictable?​

●​ Episodic vs Sequential: Do decisions depend on previous ones?​

●​ Static vs Dynamic: Does the environment change while the agent is deliberating?​

Q4. Types of Agents:

●​ Simple Reflex Agent: Uses condition-action rules.​

●​ Model-Based Agent: Uses internal state.​

●​ Goal-Based Agent: Plans actions to achieve goals.​

 ●​ Utility-Based Agent: Considers best outcome among multiple goals.​

●​ Learning Agent: Adapts through learning from past experiences.​

Q5. What is a Learning Agent?

●​ Improves its performance through experience.​

●​ Composed of four components:​

○​ Learning Element​

○​ Performance Element​

○​ Critic (gives feedback)​

○​ Problem Generator (suggests actions)​

Q6. Explain Problem Solving Agent.

●​ Starts with a goal and a current state.​

●​ Searches through possible actions to find a solution.​

●​ Returns a sequence of actions.​

Q7. How do you formulate a problem in AI?

●​ Define initial state, actions, goal test, and path cost.​

Q8. Example Problems

●​ 8-puzzle: Arrange tiles into a goal configuration.​

●​ Route finding: Find shortest path from A to B.​

●​ Missionaries and Cannibals: Transfer without violating rules.​

Module 3: Problem Solving (10 Hours)

3.1 Uninformed Search Methods

●​ Breadth-First Search (BFS): Explores all nodes at the present depth level before
moving on to the next level. It guarantees finding the shortest path in an unweighted
graph.​

●​ Depth-First Search (DFS): Explores as far down a branch as possible before

backtracking. It's less memory-intensive than BFS but may not find the shortest path.​

●​ Depth-Limited Search: A variant of DFS where a depth limit is set to avoid infinite loops
in deep graphs.​

3.2 Informed Search Methods

●​ Greedy Best First Search: Chooses the path that appears to lead most directly to the
goal, based on heuristic information.​

●​ A Search*: Combines the features of BFS and greedy search by using both the cost to
reach a node and the estimated cost to the goal.​

3.3 Local Search Algorithms

●​ Hill Climbing: Moves towards the direction of the steepest ascent, but may get stuck in
local maxima.​

●​ Simulated Annealing: A probabilistic technique that avoids local maxima by allowing

occasional uphill moves, mimicking the cooling of molten metal.​

●​ Genetic Algorithms: Use concepts of natural selection and evolution to find solutions to
optimization problems.​

3.4 Adversarial Search

●​ Game Playing: Involves two agents (players) competing against each other. Minimax
Search is used to determine the best possible move.​

●​ Min-Max Search: Explores all possible moves to determine the best strategy by
minimizing the opponent’s maximum payoff.​
 ●​ Alpha-Beta Pruning: An optimization technique to Minimax search that reduces the
number of nodes evaluated.​

Module 4: Knowledge and Reasoning (12 Hours)

4.1 Knowledge-based Agents

●​ Propositional Logic: A form of logic where statements are either true or false. It’s used
for simple deduction in AI systems.​

●​ First-Order Logic (FOL): Extends propositional logic by dealing with predicates,

quantifiers, and variables. It allows AI systems to represent more complex relationships.​

4.2 Uncertainty and Reasoning

●​ Belief Networks: Represent uncertain knowledge using nodes for variables and edges
for dependencies. Inference is performed to calculate probabilities.​

Module 5: Planning and Learning (5 Hours)

5.1 Planning

●​ State Space Search: The problem is modeled as a search through a space of states to
find the most optimal solution.​

●​ Partial Order Planning: Avoids unnecessary ordering of actions, leading to a more

flexible plan.​

5.2 Learning

●​ PAC Learning: Probably Approximately Correct Learning focuses on finding a

hypothesis that is probably close to the true concept.​

●​ Reinforcement Learning: An agent learns by receiving rewards or punishments,

improving actions based on feedback.​
Module 6: AI Applications (4 Hours)

A. Introduction to NLP

●​ Language Models: These models predict the probability of a sequence of words and
are crucial for tasks like speech recognition and text generation.​

●​ Grammars and Parsing: Understanding the structure of sentences through rules and
parsing techniques, aiding in machine understanding of natural language.​

B. Robotics

●​ Robot Hardware: Involves sensors (e.g., cameras, LIDAR), actuators (e.g., motors),
and control systems.​

●​ Robotics Problems: Includes navigation, manipulation, and interaction with humans

and environments.​

C. AI Applications

●​ Healthcare: AI systems help in diagnosing diseases, recommending treatments, and

managing patient care.​

●​ Retail: AI is used for recommendation systems, inventory management, and

personalized marketing.​

●​ Banking: AI aids in fraud detection, loan approvals, and customer service automation.​