Artificial Intelligence Lecture Slide-08

Artificial Intelligence

1. June 25, 2024 Artificial Intelligence, Lecturer #8 1
Artificial Intelligence
Lecture #08

2. June 25, 2024 Artificial Intelligence, Lecturer #8 2
Contents
Agents and Intelligent Agents
Agents And Environment
Agent Function & Agent Program
Properties of Agent
Concept of Rationality
Nature of Environment
PEAS (Performance measure, Environment,
Actuators, Sensors)
The Structure of Agent

3. June 25, 2024 Artificial Intelligence, Lecturer #8 3
Agents and Intelligent Agents
An agentis anything that can be viewed as
perceivingits environmentthrough sensorsand
actingupon that environment through actuators
An intelligent agentacts further for its own interests.

4. June 25, 2024 Artificial Intelligence, Lecturer #8 4
Example of Agents
Human agent:
•Sensors: eyes, ears, nose….
•Actuators: hands, legs, mouth, …
Robotic agent:
•Sensors: cameras and infrared range finders
•Actuators: various motors
Agents include humans, robots, thermostats, etc
Perceptions: Vision, speech reorganization, etc.

5. June 25, 2024 Artificial Intelligence, Lecturer #8 5
Agent Function & program
An agent is specified by an agent functionfthat maps seque
nces of percepts Yto actions A:
The agent programruns on the physical architectureto prod
uce f
•agent = architecture + program
“Easy” solution: table that maps every possible sequence Y
to an action A01
01
{ , ,..., }
{ , ,..., }
:
T
T
Y y y y
A a a a
f Y A




6. June 25, 2024 Artificial Intelligence, Lecturer #8 6
Agents and Environments
The agent functionmaps from percept histories
(sequences of percepts) to actions:
[f: P*A]

7. June 25, 2024 Artificial Intelligence, Lecturer #8 7
Example: A Vacuum-Cleaner Agent
A B
Percepts: location and contents, e.g., (A,dust)
•(Idealization: locations are discrete)
Actions: move, clean, do nothing:
LEFT, RIGHT, SUCK, NOP

8. June 25, 2024 Artificial Intelligence, Lecturer #8 8
Example: A Vacuum-Cleaner Agent

9. June 25, 2024 Artificial Intelligence, Lecturer #8 9
Properties of Agent
mobility:the ability of an agent to move around in an
environment.
veracity:an agent will not knowingly communicate fa
lse information
benevolence: agents do not have conflicting goals, an
d that every agent will therefore always try to do what
is asked of it
rationality: agent will act in order to achieve its goals,
and will not act in such a way as to prevent its goals b
eing achieved.
learning/adoption: agents improve performance over
time

10. June 25, 2024 Artificial Intelligence, Lecturer #8 10
Agents Vs. Objects
agents are autonomous:
agents embody stronger notion of autonomy than objects, and in
particular, they decide for themselves whether or not to perform a
n action on request from another agent
agents are smart:
capable of flexible (reactive, pro-active, social) behavior, and the
standard object model has nothing to say about such types of beh
avior
agents are active:
a multi-agent system is inherently multi-threaded, in that each ag
ent is assumed to have at least one thread of active control

11. June 25, 2024 Artificial Intelligence, Lecturer #8 11
The Concept of Rationality
What is rationalat any given time depends
on four things:
The performance measure that defines the criterio
n of success.
The agent’s prior knowledge of the environment.
The actions the agent can perform.
The agent’s percept sequence to date.

12. June 25, 2024 Artificial Intelligence, Lecturer #8 12
Rational Agents
Rational Agent:
For each possible percept sequence, a rational agent
should select an action that is expectedto maximizeit
s performance measure.
Performance measure: An objective criterion for succ
ess of an agent's behavior, given the evidence provide
d by the percept sequence.

13. June 25, 2024 Artificial Intelligence, Lecturer #8 13
Nature of Task Environment
To design a rational agent we need to specify a t
ask environment
•a problem specification for which the agent i
s a solution
PEAS: to specify a task environment
•Performance measure
•Environment
•Actuators
•Sensors

14. June 25, 2024 Artificial Intelligence, Lecturer #8 14
PEAS: Specifying an Automated
Taxi Driver
Performance measure:
•safe, fast, legal, comfortable, maximize profits
Environment:
•roads, other traffic, pedestrians, customers
Actuators:
•steering, accelerator, brake, signal, horn
Sensors:
•cameras, sonar, speedometer, GPS

15. June 25, 2024 Artificial Intelligence, Lecturer #8 15
PEAS: Another Example
Agent: Medical diagnosis system
Performance measure: Healthy patient, minimize costs.
Environment: Patient, hospital, staff
Actuators: Screen display (questions, tests, diagnoses, tre
atments, referrals)
Sensors: Keyboard (entry of symptoms, findings, patient's
answers)

16. June 25, 2024 Artificial Intelligence, Lecturer #8 16
Properties of Task Environment
Fullyobservable:(vs.partiallyobservable):Anagent'ssenso
rsgiveitaccesstothecompletestateoftheenvironmentateac
hpointintime.
Deterministic(vs. stochastic): The next state of the environme
nt 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

17. June 25, 2024 Artificial Intelligence, Lecturer #8 17
Episodic(vs. sequential): The agent's experience is divided in
to atomic episodesduring which the agent perceives and then
performs a single action, and the choice of action in each episo
de depends only on the episode itself.
Episodicenvironments are simpler from the agent developer’s
perspective because the agent can decide what action to perfor
m based only on the current episode —it need not reason abou
t the interactions between this and future episodes.
Properties of Task Environment

18. June 25, 2024 Artificial Intelligence, Lecturer #8 18
A staticenvironment is one that can be assumed to
remain unchanged except by the performance of a
ctions by the agent
A dynamicenvironment is one that has other proc
esses operating on it, and which hence changes in
ways beyond the agent’s control.
The physical world is a highly dynamic environment.
Properties of Task Environment
Static Vs. Dynamic

19. June 25, 2024 Artificial Intelligence, Lecturer #8 19
Discrete vs. continuous
An environment is discreteif there are a finite nu
mber of distinct statesin the environment and a di
screte set of perceptsand actions.
The game of chess is an example of a discrete environment,
automated taxi driver is a continuous-state and continuous-t
ime problem.
Discrete environments could in principle be handl
ed by a kind of “lookup table”
Properties of Task Environment

20. June 25, 2024 Artificial Intelligence, Lecturer #8 20
Structure Of Agent
Goals
•Given a PEAS task environment
•construct agent function f,
•design an agent program that implements fon a particul
ar architecture
•Agent= Architecture +program.
Agent Architecture:
•Computing device with physicalsensorand actuator.
•Makes the percept from the sensors and make it available
to the program.
•Runs the program
•Feeds the program action choices to the actuators.

21. June 25, 2024 Artificial Intelligence, Lecturer #8 21
Recommended Textbooks
[Negnevitsky,2001]M.Negnevitsky“ArtificialIntelligence:Aguideto
IntelligentSystems”,PearsonEducationLimited,England,2002.
[Russel,2003]S.RussellandP.NorvigArtificialIntelligence:AModern
ApproachPrenticeHall,2003,SecondEdition
[Patterson,1990]D.W.Patterson,“IntroductiontoArtificialIntelligence
andExpertSystems”,Prentice-HallInc.,EnglewoodCliffs,N.J,USA,19
90.
[Minsky,1974]M.Minsky“AFrameworkforRepresentingKnowledge”,
MIT-AILaboratoryMemo306,1974.
[Hubel,1995]DavidH.Hubel,“Eye,Brain,andVision”
[Ballard,1982]D.H.BallardandC.M.Brown,“ComputerVision”,
PrenticeHall,1982.

22. June 25, 2024 Artificial Intelligence, Lecturer #8 22
Other References
http://en.wikipedia.org/wiki/Intelligent_agent
http://aima.eecs.berkeley.edu/slides-ppt/m2-agents.ppt#3
http://www.cs.cmu.edu/~sandholm/cs15-381/Agents.ppt#2

23. June 25, 2024 Artificial Intelligence, Lecturer #8 23
End of Presentation
Thanks to all !!!