1. CSC 9010 Spring 2011. Paula Matuszek Intelligent Agents Overview Slides based in part on Hwee Tou Ng, aima.eecs.berkeley.edu/slides-ppt, which are in turn based on Russell, aima.eecs.berkeley.edu/slides- pdf.

2. 2 CSC 9010 Spring 2011. Paula Matuszek Agents An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators Human agent: eyes, ears, and other organs for sensors; hands, legs, mouth, and other body parts for actuators Robotic agent: cameras and infrared range finders for sensors; various motors for actuators Scooter: touch and rotation sensors; wheels

3. 3 CSC 9010 Spring 2011. Paula Matuszek 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 = architecture + program

4. 4 CSC 9010 Spring 2011. Paula Matuszek Vacuum-cleaner world Percepts: location and contents, e.g., [A,Dirty] Actions: Left, Right, Suck, NoOp

5. 5 CSC 9010 Spring 2011. Paula Matuszek A vacuum-cleaner agent Percept sequenceAction [A,Clean]Right [A, Dirty]Suck [B, Clean]Left [B, Dirty]Suck [A, Clean],[A, Clean]Right [A, Clean],[A, Dirty]Suck ……

6. 6 CSC 9010 Spring 2011. Paula Matuszek Rational agents 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.

7. 7 CSC 9010 Spring 2011. Paula Matuszek Rational agents 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. Thus, rationality depends on –performance measure that defines success –prior knowledge of the environment –actions the agent can perform –percept sequence to date

8. 8 CSC 9010 Spring 2011. Paula Matuszek Rational agents 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)

9. 9 CSC 9010 Spring 2011. Paula Matuszek Task Environment An agent operates within some task environment, not in a blank world. This environment includes: –what the agent is trying to do –what resources it has to do it The nature of the environment affects how we design an appropriate agent. 9

10. 10 CSC 9010 Spring 2011. Paula Matuszek PEAS: Specifying an Agent's World The task environment for an agent can be completely specified by defining four things: –Performance measure: How do we assess whether we are doing the right thing? –Environment: What is the world we are in? –Actuators: How do we affect the world we are in? –Sensors: How do we perceive the world we are in? This PEAS specification gives us the information we need to design a rational agent.

11. 11 CSC 9010 Spring 2011. Paula Matuszek PEAS: Taxi Driver 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. 12 CSC 9010 Spring 2011. Paula Matuszek PEAS: Agent: Lego Scooter Performance: number of seconds it keeps moving without getting stuck Environment: Flat surface with objects but no dropoffs. Actuators: Motors which turn two wheels Sensors: touch sensors, rotation sensors 12

13. 13 CSC 9010 Spring 2011. Paula Matuszek PEAS Agent: Medical diagnosis system –Performance measure: –Environment: –Actuators: –Sensors:

14. 14 CSC 9010 Spring 2011. Paula Matuszek PEAS 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)

15. 15 CSC 9010 Spring 2011. Paula Matuszek PEAS Agent: Interactive English tutor Performance measure: Environment: Actuators: Sensors:

16. 16 CSC 9010 Spring 2011. Paula Matuszek PEAS Agent: Interactive English tutor Performance measure: Maximize student's score on test Environment: Set of students Actuators: Screen display (exercises, suggestions, corrections) Sensors: Keyboard

17. 17 CSC 9010 Spring 2011. Paula Matuszek Agent functions and programs An agent is completely specified by the agent function mapping percept sequences to actions A rational agent function maximizes the average performance for a given environment class Aim: find a way to implement the rational agent function concisely

18. 18 CSC 9010 Spring 2011. Paula Matuszek Agent types Four basic types in order of increasing generality: Simple reflex agents Model-based reflex agents Goal-based agents Utility-based agents

19. 19 CSC 9010 Spring 2011. Paula Matuszek Simple reflex agents

20. 20 CSC 9010 Spring 2011. Paula Matuszek Simple reflex Vacuum Agent function REFLEX-VACUUM-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 Observe the world, choose an action, implement action, done. Problems if environment is not fully-observable. Depending on performance metric, may be inefficient. Even this simple agent can have a knowledge base, in the form of condition-action rules.

21. 21 CSC 9010 Spring 2011. Paula Matuszek Model-Based Agents Suppose moving has a cost? If a square stays clean once it is clean, then this algorithm will be extremely inefficient. A very simple improvement would be –Record when we have cleaned a square –Don’t go back once we have cleaned both. We have built a very simple model.

22. 22 CSC 9010 Spring 2011. Paula Matuszek Reflex Agents with State

23. 23 CSC 9010 Spring 2011. Paula Matuszek Reflex Agents with State  More complex agent with model: a square can get dirty again. Function REFLEX_VACUUM_AGENT_WITH_STATE ([location, status]) returns an action. last-cleaned-A and last-cleaned-B initially declared = 100. Increment last-cleaned-A and last-cleaned-B. if status == Dirty then return Suck if location == A then set last-cleaned-A to 0 if last-cleaned-B > 3 then return right else no-op else set last-cleaned-B to 0 if last-cleaned-A > 3 then return left else no-op  The value we check last-cleaned against could be modified.  Could track how often we find dirt to compute value

24. 24 CSC 9010 Spring 2011. Paula Matuszek Model-Based = Reflex Plus State Maintain an internal model of the state of the environment Over time update state using world knowledge –How the world changes –How actions affect the world Agent can operate more efficiently More effective than a simple reflex agent for partially observable environments May use a KB for both condition-action rules and what world/actions do.

25. 25 CSC 9010 Spring 2011. Paula Matuszek Goal-based agents

26. 26 CSC 9010 Spring 2011. Paula Matuszek Goal-Based Agent Agent has some information about desirable situations Needed when a single action cannot reach desired outcome Therefore performance measure needs to take into account "the future". Typical model for search and planning.

27. 27 CSC 9010 Spring 2011. Paula Matuszek Utility-based agents

28. 28 CSC 9010 Spring 2011. Paula Matuszek Utility-Based Agents Possibly more than one goal, or more than one way to reach it Some are better, more desirable than others There is a utility function which captures this notion of "better". Utility function maps a state or sequence of states onto a metric.

29. 29 CSC 9010 Spring 2011. Paula Matuszek Learning agents

30. 30 CSC 9010 Spring 2011. Paula Matuszek Learning Agents All agents have methods for selection actions. Learning agents can modify these methods. Performance element: any of the previously described agents Learning element: makes changes to actions Critic: evaluates actions, gives feedback to learning element Problem generator: suggests actions

31. 31 CSC 9010 Spring 2011. Paula Matuszek Summary We can view most intelligent systems as agents. An agent operates in a world which can be described by its Performance measure, Environment, Actuators, and Sensors. A rational agent chooses actions which maximize its performance measure, given the information it has. Agents range in complexity from simple reflex agents to complex utility-based and learning agents.