Artificial Intelligence - Important Questions with Answers
1. What is Artificial Intelligence?
Artificial Intelligence (AI) is the simulation of human intelligence processes by machines, especially computer
systems. These processes include learning (the acquisition of information and rules for using the
information), reasoning (using the rules to reach approximate or definite conclusions), and self-correction. AI
applications include expert systems, natural language processing (NLP), speech recognition, and machine
vision.
2. Explain reflex and model-based reflex agents. What's the difference?
A simple reflex agent acts only on the basis of the current percept, ignoring the rest of the percept history. It
follows condition-action rules. For example, if a sensor detects an obstacle, the robot turns.
A model-based reflex agent, on the other hand, maintains some internal state based on the percept history
and the model of the environment. This allows it to handle partial observability better than simple reflex
agents. The key difference lies in the use of internal state.
3. Explain problem-solving agents and their types.
A problem-solving agent is a type of goal-based agent that decides what to do by finding sequences of
actions that lead to desirable states. These agents are composed of components such as a goal formulation
module, problem formulation module, and search module. They include types such as simple problem-solving
agents, planning agents, and search agents that use strategies like breadth-first search (BFS), depth-first
search (DFS), etc.
4. Compare breadth-first search and depth-first search strategies.
Breadth-First Search (BFS) explores all nodes at the present depth before moving on to the nodes at the next
depth level. It is complete, optimal (if path cost is a non-decreasing function of depth), and uses more
memory.
Artificial Intelligence - Important Questions with Answers
Depth-First Search (DFS) explores as far as possible along each branch before backtracking. It is more
memory efficient, but can get stuck in loops and may not find the shortest path.
5. Explain hill climbing and simulated annealing search with comparison.
Hill climbing is a local search algorithm that continuously moves in the direction of increasing value (uphill) to
find the peak of the mountain (maximum value). However, it may get stuck in local maxima. Simulated
annealing, inspired by metallurgy, allows occasional moves to worse states to escape local maxima. Over
time, the probability of such moves decreases. This makes simulated annealing better at avoiding local
optima compared to hill climbing.
6. What are local and global search strategies? Explain with an example.
Global search strategies such as BFS or DFS search through the entire search space and are good for
discrete spaces. Local search strategies operate on a single current state and move to neighbors, useful in
large or continuous spaces. For example, in optimizing neural network weights, local search (like gradient
descent) works better because evaluating the whole state space is infeasible.
7. Define an intelligent agent and its types.
An intelligent agent is an entity that perceives its environment through sensors and acts upon that
environment through actuators. Types include:
1. Simple reflex agents
2. Model-based reflex agents
3. Goal-based agents
4. Utility-based agents
5. Learning agents
Artificial Intelligence - Important Questions with Answers
8. Applications of Artificial Intelligence (AI)
AI is applied across various domains:
- Healthcare: Diagnosis, robotic surgeries, drug discovery
- Automotive: Self-driving cars
- Finance: Fraud detection, algorithmic trading
- Education: Personalized learning
- Agriculture: Crop disease detection
- Defense: Autonomous drones, surveillance
9. Explain alpha-beta pruning with examples.
Alpha-beta pruning is an optimization technique for the minimax algorithm that eliminates branches in a game
tree that don't affect the final decision. Alpha represents the maximum value that the maximizer is assured of,
and beta represents the minimum value that the minimizer is assured of. By pruning branches where alpha
>= beta, computation is reduced.
10. Explain Bayesian networks and probabilistic reasoning in AI.
Bayesian networks are probabilistic graphical models that represent a set of variables and their conditional
dependencies via a directed acyclic graph (DAG). They are useful for reasoning under uncertainty.
Probabilistic reasoning allows agents to make informed decisions even when information is incomplete or
uncertain. Applications include medical diagnosis, spam detection, and weather forecasting.
