Lecture 2
Download as PPT, PDF0 likes738 views
This document discusses intelligent agents and their environments. It defines an agent as anything that perceives its environment and acts upon it. Rational agents are those that select actions expected to maximize their performance based on perceptions. An agent's task environment consists of its performance measure, environment, actuators, and sensors. Environment types include fully/partially observable, deterministic/stochastic, and single/multi-agent. Four basic agent types are described: simple reflex agents, model-based reflex agents, goal-based agents, and utility-based agents. Learning agents use feedback to improve performance over time. The document provides examples of agents and discusses their design considerations based on task environment properties.
![Agents and environments
• The agent function maps from percept histories
to actions:
[f: P* A]
• The agent program runs on the physical
architecture to produce f.
• agent comprised of : architecture + program](https://image.slidesharecdn.com/lecture2-120919112122-phpapp01/85/Lecture-2-4-320.jpg)
![Vacuum-cleaner world
• Percepts: location and contents, e.g., [A,Dirty]
• Actions: Left, Right, Suck, NoOp](https://image.slidesharecdn.com/lecture2-120919112122-phpapp01/85/Lecture-2-5-320.jpg)
![A vacuum-cleaner agent
Percept Sequence Action
[A, Clean] Right
[A,Dirty] Suck
[B, Clean] Left
[B, Dirty] Suck
[A, Clean], [A, Clean] Right
[A,Clean], [A,Dirty] Suck
…. …..
[A, Clean], [A, Clean], [A, Clean] Right
[A, Clean], [A, Clean], [A, Dirty] Suck
….. ….](https://image.slidesharecdn.com/lecture2-120919112122-phpapp01/85/Lecture-2-6-320.jpg)
![Agent program for a vacuum-
cleaner agent
Function Reflex_Va_Agent([location,
status]) return an action
If status = Dirty then return Suck
Else if location = A then return Right
Else if location = B then return Left
Good only if environment is fully observable otherwise disaster.
For example if location sensor is off then problem? [dirty] [clean]](https://image.slidesharecdn.com/lecture2-120919112122-phpapp01/85/Lecture-2-22-320.jpg)
Lecture 2
- 2. Outline • Agents and environments • Rationality • Task Environment • Environment types • Agent types • Examples
- 3. Agents An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators. Human agent: Sensors: eyes, ears, and other organs for sensors; Actuators: hands, legs, mouth, and other body parts. Robotic agent: Sensors: cameras and infrared range finders Actuator: various motors for actuators
- 4. Agents and environments • The agent function maps from percept histories to actions: [f: P* A] • The agent program runs on the physical architecture to produce f. • agent comprised of : architecture + program
- 5. Vacuum-cleaner world • Percepts: location and contents, e.g., [A,Dirty] • Actions: Left, Right, Suck, NoOp
- 6. A vacuum-cleaner agent Percept Sequence Action [A, Clean] Right [A,Dirty] Suck [B, Clean] Left [B, Dirty] Suck [A, Clean], [A, Clean] Right [A,Clean], [A,Dirty] Suck …. ….. [A, Clean], [A, Clean], [A, Clean] Right [A, Clean], [A, Clean], [A, Dirty] Suck ….. ….
- 7. Rational agents (1/3) • An agent should strive to "do the right thing", based on what it can perceive and the actions it can perform. The right action is the one that will cause the agent to be most successful. • Performance measure: An objective criterion for success of an agent's behavior. • E.g., performance measure of a vacuum-cleaner agent could be amount of dirt cleaned up, amount of time taken, amount of electricity consumed, amount of noise generated, etc.
- 8. Rational agents (2/3) • Rational Agent: For each possible percept sequence, • a rational agent should select an action that is expected to maximize its performance measure, • given the evidence provided by the percept sequence and whatever built-in knowledge the agent has.
- 9. Rational agents (3/3) • Rationality is distinct from omniscience (all-knowing with infinite knowledge). • Agents can perform actions in order to modify future percepts so as to obtain useful information (information gathering, exploration). • An agent is autonomous if its behavior is determined by its own experience (with ability to learn and adapt).
- 10. Rational agents (1/5) • What is rational depends on: – Performance measure - The performance measure that defines the criterion of success – Environment - The agents prior knowledge of the environment – Actuators - The actions that the agent can perform – Sensors - The agent’s percept sequence to date • We’ll call all this the Task Environment (PEAS)
- 11. Task environment (2/5) Must first specify the setting for intelligent agent design. •Consider, e.g., the task of designing an automated taxi driver: – Performance measure: Safe, fast, legal, comfortable trip, maximize profits. – Environment: Roads, other traffic, pedestrians, customers. – Actuators: Steering wheel, accelerator, brake, signal, horn. – Sensors: Cameras, sonar, speedometer, GPS, odometer, engine sensors, keyboard.
- 12. Task environment (3/5) Agent: Medical diagnosis system • Performance measure: Healthy patient, minimize costs, lawsuits. • Environment: Patient, hospital, staff. • Actuators: Screen display (questions, tests, diagnoses, treatments, referrals). • Sensors: Keyboard (entry of symptoms, findings, patient's answers).
- 13. Task environment (4/5) Agent: Part-picking robot • Performance measure: Percentage of parts in correct bins • Environment: Conveyor belt with parts, bins • Actuators: Jointed arm and hand • Sensors: Camera, joint angle sensors
- 14. Task environment (5/5) Agent: Interactive English tutor • Performance measure: Maximize student's score on test • Environment: Set of students • Actuators: Screen display (exercises, suggestions, corrections) • Sensors: Keyboard
- 15. Environment types • Fully observable (vs. partially observable): An agent's sensors give it access to the complete state of the environment at each point in time. • Deterministic (vs. stochastic): The next state of the environment is completely determined by the current state and the action executed by the agent. (If the environment is deterministic except for the actions of other agents, then the environment is strategic). • Episodic (vs. sequential): The agent's experience is divided into atomic "episodes" (each episode consists of the agent perceiving and then performing a single action), and the choice of action in each episode depends only on the episode itself.
- 16. Environment types • Static (vs. dynamic): The environment is unchanged while an agent is deliberating. (The environment is semidynamic if the environment itself does not change with the passage of time but the agent's performance score does). • Discrete (vs. continuous): A limited number of distinct, clearly defined percepts and actions. • Single agent (vs. multiagent): An agent operating by itself in an environment.
- 17. Environment types Chess with Chess without Taxi driving a clock a clock Fully observable Yes Yes No Deterministic Strategic Strategic No Episodic No No No Static Semi Yes No Discrete Yes Yes No Single agent No No No • The environment type largely determines the agent design. • The real world is (of course) partially observable, stochastic, sequential, dynamic, continuous, multi-agent.
- 18. Agent functions and programs • An agent is completely specified by the agent function mapping percept sequences to actions • One agent function (or a small equivalence class) is rational. • Aim: find a way to implement the rational agent function concisely.
- 19. Table-lookup agent Function Table_Driven_Agent(percept) return an action Static: percepts, a sequence initially empty table: a table of actions, indexed by percept sequence, fully specified append percept to the end of percepts action ← LookUp(percepts, table) return action Drawbacks: – Huge table (10250,000,000,000 entries for an hour driving) – Take a long time to build the table – No autonomy – Even with learning, need a long time to learn the table entries
- 20. Agent types Four basic types: • Simple reflex agents • Model-based reflex agents • Goal-based agents • Utility-based agents
- 21. Simple reflex agents Sensors Agent What the world is now like? Environment Condition-action rule What action should I do now? Actuator
- 22. Agent program for a vacuum- cleaner agent Function Reflex_Va_Agent([location, status]) return an action If status = Dirty then return Suck Else if location = A then return Right Else if location = B then return Left Good only if environment is fully observable otherwise disaster. For example if location sensor is off then problem? [dirty] [clean]
- 23. Model-based reflex agents Keep track of the world it can’t see now. Consider lane change. state Sensor How does the world evolve? What the world is now like? Environment What my action do? Condition-action rule What action should I do now? Actuator Agent
- 24. Model-based reflex agents Function Reflex_agent_with_state(percept) return an action Static: state, a description of the current world rules, a set of condition action rules action, the most recent action, initially none state ← Update-State (state, action, percept) rule ← Rule-Match (state, rules) action ← Rule-action (rule) return action Knowing about the current environment is not always enough to decide what to do For example in the junction, left, right, go straight
- 25. Goal-based agents Choose actions so as to achieve a (given or computed) goal. state Sensor How does the world evolve? What the world is now like? Environment What my action do? What it will be if take action A? Goal What action should I do now? Actuator Agent
- 26. Utility-based agentsbut some are Goal alone is not enough. There can be many way to achieve goal quicker, safer, more reliable or cheaper than other. state Sensor How does the world evolve? What the world is now like? Environment What my action do? What it will be if take action A? Utility How happy I will be in that state? What action should I do now? Actuator Agent
- 27. Learning agents Performance Standard Sensor Critic feedback Environment changes Learning Performance Element Element Knowledge Learning goals Problem Generator Actuator Agent
- 28. Experience the agent You can use Webots simulator for 30 days with full functionality for free Get it from: http://www.cyberbotics.com/download Try some agent and see the controller Worm Rat in Maze Pioneer quadruped Sumo
- 29. Thank You