Artificial Intelligence Lecture Slide-08

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

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

Agents and Intelligent Agents
An agent is anything that can be viewed as
 perceiving its environment through sensors and
 acting upon that environment through actuators
 An intelligent agent acts further for its own interests.

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.

Agent Function & program
 An agent is specified by an agent function f that maps seque
nces of percepts Y to actions A:
 The agent program runs on the physical architecture to prod
uce f
• agent = architecture + program
 “Easy” solution: table that maps every possible sequence Y
to an action A
0 1
0 1
{ , ,..., }
{ , ,..., }
:
T
T
Y y y y
A a a a
f Y A




Agents and Environments
 The agent function maps from percept histories
(sequences of percepts) to actions:
[f: P*  A]

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

Example: A Vacuum-Cleaner Agent

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

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

The Concept of Rationality
 What is rational at 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.

Rational Agents
 Rational Agent:
For each possible percept sequence, a rational agent
should select an action that is expected to maximize it
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.

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

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

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)

Properties of Task Environment
 Fully observable :(vs. partially observable): An agent's senso
rs give it access to the complete state of the environment at eac
h point in time.
 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

 Episodic (vs. sequential): The agent's experience is divided in
to atomic episodes during 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.
 Episodic environments 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.

 A static environment is one that can be assumed to
remain unchanged except by the performance of a
ctions by the agent
 A dynamic environment 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.
Static Vs. Dynamic

Discrete vs. continuous
An environment is discrete if there are a finite nu
mber of distinct states in the environment and a di
screte set of percepts and 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”

Structure Of Agent
 Goals
• Given a PEAS task environment
• construct agent function f,
• design an agent program that implements f on a particul
ar architecture
• Agent= Architecture +program.
 Agent Architecture:
• Computing device with physical sensor and 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.

Recommended Textbooks
 [Negnevitsky, 2001] M. Negnevitsky “ Artificial Intelligence: A guide to
Intelligent Systems”, Pearson Education Limited, England, 2002.
 [Russel, 2003] S. Russell and P. Norvig Artificial Intelligence: A Modern
Approach Prentice Hall, 2003, Second Edition
 [Patterson, 1990] D. W. Patterson, “Introduction to Artificial Intelligence
and Expert Systems”, Prentice-Hall Inc., Englewood Cliffs, N.J, USA, 19
90.
 [Minsky, 1974] M. Minsky “A Framework for Representing Knowledge”,
MIT-AI Laboratory Memo 306, 1974.
 [Hubel, 1995] David H. Hubel, “Eye, Brain, and Vision”
 [Ballard, 1982] D. H. Ballard and C. M. Brown, “Computer Vision”,
Prentice Hall, 1982.

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

End of Presentation
Thanks to all !!!

Artificial Intelligence Lecture Slide-08

More Related Content

Similar to Artificial Intelligence Lecture Slide-08 (20)

More from asmshafi1 (11)

Recently uploaded (20)

Artificial Intelligence Lecture Slide-08