Quiz 1 Prep - AI Concepts (PDF)

**Conceptual Questions** **1. Definitions and Understanding** - **Question:** What is AI? - **Answer:** The art of creating machines that perform functions that require intelligence when performed by humans - **Question:** What Turing Test is used for? - **Answer:** used to check if...

**Conceptual Questions** **1. Definitions and Understanding** - **Question:** What is AI? - **Answer:** The art of creating machines that perform functions that require intelligence when performed by humans - **Question:** What Turing Test is used for? - **Answer:** used to check if the AI is working properly or not. - **Question:** Define the terms \"agent\" and \"environment\" in the context of artificial intelligence. - **Answer:** An *agent* is an entity that perceives its environment through sensors and acts upon it using actuators to achieve specific goals. The *environment* is the external context or world in which an agent operates and interacts. **2. Characteristics of Agents** - **Question:** List and explain the key properties of intelligent agents. - **Answer:** Key properties include: - **Autonomy:** Ability to operate without human intervention. - **Situatedness:** Agent receives some form of sensory input from its environment and it performs some action that changes its environment in some way. - **Adaptivity:** Reacting flexibly to changes in its environment. - **Reactivity:** Responding to changes in the environment. - **Proactiveness:** Taking initiative to achieve goals. - **Sociability:** Interacting with other agents or humans (peer to peer). - **Question:** List the types of agent program with examples: - **Answer:** - **Simple Reflex Agents-** Act based on the current percept only. **Example**: A thermostat. - **Model-Based Reflex Agent -** Use a model to track unobservable parts of the environment. **Example**: A robot vacuum cleaner. - **Goal-Based Agent -**Take actions to achieve specific goals. **Example**: GPS navigation. - **Utility-Based Agents -** Use a utility function to maximize desirable outcomes.(جنه satisfaction ) **Example**: Movie recommendation system. - A *simple reflex agent* acts based on current percepts without considering the future (e.g., a thermostat turning heating on/off based on temperature). - A *goal-based agent* acts to achieve specific goals, considering future states (e.g., a robot planning a path to deliver an item). **3. Types of Environments:** - **Question:** Classify the following environments: - **A chess game:** - Fully observable, deterministic, static, discrete. - **A robot navigating a room with movable obstacles:** - Partially observable, stochastic, dynamic, continuous. - **An online recommendation system:** - Partially observable, stochastic, dynamic, discrete. **Scenario-Based Questions** **4. Agent Design** - **Question:** Identify percepts, actions, performance measure and environment of Tic-Tac-Toe. - **Answer:** - **Percepts:** Current board state. - **Actions:** Place X or O in an empty cell. - **Performance measure:** Win the game, minimize losses. - **Environment:** Fully observable, deterministic, static, discrete. **5. Environment Classification** - **Question:** Consider a self-driving car and classify its environment: - **Answer:** - Observability: Partially observable (limited by sensors). - Determinism: Stochastic (uncertainties like pedestrian behavior). - Static/Dynamic: Dynamic (environment changes over time). - Discrete/Continuous: Continuous (e.g., road positions). **6. Agent Performance** - **Question:** Define the performance measure for a warehouse sorting agent. - **Answer:** - Accuracy in placing packages. - Speed of sorting. - Minimizing errors like misplacement. - Adapting to varying loads efficiently. ***[READ AND UNDERSTAND :]*** **[Environment Types]** **1. Static vs Dynamic** These describe whether the environment changes over time. **Dynamic Environment:** - The environment changes while the agent is deciding what to do. - The agent must act quickly or keep updating its decision to respond to changes. - Example: A self-driving car must react to moving vehicles, pedestrians, and traffic lights while making decisions. **Static Environment:** - The environment does not change while the agent decides on an action. - The agent can take its time to think without worrying about time or external changes. - Example: Solving a puzzle like Sudoku. The puzzle doesn't change while you work on it. **Semi-Dynamic Environment:** - The environment itself doesn't change, but the agent's performance score or preferences may change with time. - Example: Online shopping with a limited-time discount. The website (environment) doesn't change, but the agent (shopper) must act quickly as the offer may expire. **2. Discrete vs Continuous** These describe whether the environment has a limited or unlimited number of actions and states. **Discrete Environment:** - Has a limited number of clear options or states. - The agent can easily count and identify all possible actions and outcomes. - Example: - Chess: There are specific, finite moves the player can make on the board. - Elevator Control: The floors are discrete states, and the elevator can only move up or down between them. **Continuous Environment:** - Has an infinite range of possibilities for actions or states. - The agent must deal with complex variables that can change gradually. - Example: - Driving a Taxi: Speed, steering, and location continuously vary. The agent must handle smooth, ongoing changes. *Summary* - Static/Dynamic deals with whether the environment changes over time. - Discrete/Continuous deals with whether the environment has clear, limited options or infinite possibilities. - This classification helps agents adapt their behavior to different situations.