Scribd
Upload
4 views

Week_2_A_Intelligent Agents

This document discusses the concepts of intelligent agents and their environments, focusing on the definition of rational agents and the factors influencing their performance. It introduces the PEAS framework for specifying task environments and outlines various properties such as observability, determinism, and agent interactions. Additionally, it highlights the importance of correctly defining performance measures to avoid issues like the King Midas problem.

Uploaded by

coham90828
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
4 views

Week_2_A_Intelligent Agents

This document discusses the concepts of intelligent agents and their environments, focusing on the definition of rational agents and the factors influencing their performance. It introduces the PEAS framework for specifying task environments and outlines various properties such as observability, determinism, and agent interactions. Additionally, it highlights the importance of correctly defining performance measures to avoid issues like the King Midas problem.

Uploaded by

coham90828
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 23

Intelligent

Agent
WEEK – 2, PART – A
ABDUL BASEER BURIRO
This Chapter (Week) `ll Cover
• Agents
• Environments
• The Nature of Environments
• The structure of agents
• The Components of Agent Programs
Rational Agent – A Quick Recall
An agent is anything that can be
viewed as:
◦ Perceive its environment through sensors
◦ Act upon that environment through
actuators

• Rational  doing the right thing under


given circumstances

Perceive Think Act


Logically Then …
• Some agents perform better than
others
• Such performances depend on the
nature of the environment
Environment

Perceive Think Act


The Environment
•The environment could be everything
—the entire universe!
•In practice it is a part of the universe
whose state we care about
• Percept  the content an agent’s
sensors are perceiving
• Percept sequence is the complete
history of everything the agent has
ever perceived
The Environment
• Desired change in the environment 
Performance
• A desirable state  agent has
performed well
• A successful agent  intelligent agent
Agent Function
• An agent’s behavior (Mathematically)
• Maps a given percept sequence to an
action
• An agent program (code) implements We aim to develop
the agent function

Built-in
Percepts Actions
Knowledge
Agent Function – Example

Note that the table is of unbounded size unless there is a restriction on the
length of possible percept sequences.
Good Behavior: Rationality
Concept
•An agent that does the right things is a rational agent
•Right thing? (how to measure)
•Desirable sequence (Performance measure)
• Performance measure needs to be specified correctly
• There is a possibility that the wrong purpose is being put into the machine —
precisely the King Midas problem
King Midas Problem
•Many cultures have myths of humans who ask
for something.
•Invariably, they get what they literally ask for,
and then regret it.
•The third wish, if there is one, is to undo the
first two. The King Midas problem
• Midas, a legendary King in Greek mythology,
asked that everything he touched should turn
to gold, but then regretted it after touching his
food, drink, and family members
Rationality
•Rationality at a given time depends on four things:
1. The performance measure that defines the criterion of success
2. The agent’s prior Knowledge of the environment
3. The actions that agent can perform
4. The agent’s percept sequence to date

• 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.
Is it a Rational Agent? – Example

• Under some circumstances the agent is indeed rational; its expected


performance is at least as good as any other agent’s
• But under different circumstances, it would be irrational, e.g., oscillating
needlessly back and forth
The Nature of Environments
•The task environments are essentially the “problems” to which rational agents
are the “solutions”
•The nature of the task environment directly affects the appropriate design for
the agent program
Specifying the Task Environment

PEAS (Performance, Environment, Actuators, Sensors)


PEAS Description – Example
Properties of Task Environments
•Observable Vs Partially Observable
• Deterministic vs Non-deterministic
•Episodic vs Sequential
•Single- vs Multi-agent
•Static vs Dynamic
•Known vs Unknown
Fully, Partially, & Un Observable
Single-Agent vs. Multi-Agent
The key distinction
is whether B’s
behavior is best
described as
maximizing a
performance
measure whose
value depends on
agent A’s behavior.
Deterministic vs. Non-
Deterministic
Episodic vs. Sequential
Static vs. Dynamic
Discrete vs. Continuous
Task Examples and their
Characteristics

You might also like