CHAPTER FOUR

INTELLIGENT AGENTS IN AI

BY Abdela Aman. 09/30/2021

 INTRODUCTION
2

 An AI system can be defined as the study of the rational

agent and its environment.

 An agent can be anything that perceive its environment

through sensors and

 act upon that environment through actuators.

 An AI agent can have mental properties such as

knowledge, belief, intention, etc.

BY Abdela Aman. 09/30/2021

 What is an Agent?
3

 An Agent runs in the cycle of perceiving, thinking,

and acting.
 An agent gets percepts one at a time, and maps this percept

sequence to actions.
 Hence the world around us is full of agents such as

thermostat, cell phone, camera, and even we are also agents.

 Before moving forward, we should first know about

sensors, effectors, and actuators

.
BY Abdela Aman. 09/30/2021
 Agents….
4
 An agent can be:
 Human-Agent: A human agent has eyes, ears, and other organs
which work for sensors and hand, legs, vocal tract work for
actuators.
 Robotic Agent: A robotic agent can have cameras, infrared range
finder, NLP for sensors and various motors for actuators.
 Software Agent: Software agent can have keystrokes, file
contents as sensory input and act on those inputs and display
output on the screen.

BY Abdela Aman. 09/30/2021

 EXAMPLES5 OF AGENTS

BY Abdela Aman. 09/30/2021

 EXAMPLES OF AGENTS
6

BY Abdela Aman. 09/30/2021

 Agents…
7

BY Abdela Aman. 09/30/2021

 Agents and Environment
8

 An agent “perceives its environment through sensors and acts

upon that environment through actuators.”

 An agent's choice of action can depend on the entire history of

percepts observed previously, but not on anything it has not

perceived.
 An agent can be seen as a mapping between percept sequences

and actions.
Agent : Percept ∗ Action

BY Abdela Aman. 09/30/2021

 Structure of an AI Agent
9

 The less an agents relies on its built-in knowledge, as opposed to

the current percept sequence, the more autonomous it is.

 The agent program runs on the physical architecture to produce

function.
 The task of AI is to design an agent program which implements

the agent function.

 A Rational agent is an agent whose acts try to maximize some

performance measure.

BY Abdela Aman. 09/30/2021

 Structure of an AI Agent…
10

 The structure of an intelligent agent is a combination of

architecture and agent program.
 Agent = Architecture + Agent program  
Architecture: Architecture is machinery that an AI agent
executes on.
Agent Function: Agent function is used to map a percept to
an action.
 f:P* → A  
Agent program: Agent program is an implementation of
agent function.
An agent program executes on the physical architecture to
produce function f.
BY Abdela Aman. 09/30/2021
 Agents and environment
11

BY Abdela Aman. 09/30/2021

 SENSOR AND EFFECTORS
12

 Sensor: Sensor is a device which detects the change in the

environment and sends the information to other electronic
devices.
 An agent observes its environment through sensors.
 Actuators: Actuators are the component of machines that
converts energy into motion.
 The actuators are only responsible for moving and controlling a system.
 An actuator can be an electric motor, gears, rails, etc.
 Effectors: Effectors are the devices which affect the
environment.
 Effectors can be legs, wheels, arms, fingers, wings, fins, and display screen.

BY Abdela Aman. 09/30/2021

 Intelligent Agents:
13
 An intelligent agent is an autonomous entity which act upon
an environment using sensors and actuators for achieving
goals.
 An intelligent agent may learn from the environment to
achieve their goals.
 Following are the main four rules for an AI agent:
Rule 1: An AI agent must have the ability to perceive the
environment.
Rule 2: The observation must be used to make decisions.
Rule 3: Decision should result in an action.
Rule 4: The action taken by an AI agent must be a rational
action.
BY Abdela Aman. 09/30/2021
 Intelligent Agents:
14

BY Abdela Aman. 09/30/2021

 Intelligent Agents:
15

BY Abdela Aman. 09/30/2021

 Rational Agent:
16

 A rational agent is an agent which has clear preference, models

uncertainty, and acts in a way to maximize its performance measure

with all possible actions.
 A rational agent is said to perform the right things.

 For an AI agent, the rational action is most important because in AI

reinforcement learning algorithm, for each best possible action, agent

gets the positive reward and for each wrong action, an agent gets a
negative reward.
 Note: Rational agents in AI are very similar to intelligent
agents.

BY Abdela Aman. 09/30/2021

 Rational Agent:
17

BY Abdela Aman. 09/30/2021

 Rational Agent:
18

BY Abdela Aman. 09/30/2021

 Rationality:
19

 The rationality of an agent is measured by its performance

measure.
 Rationality can be judged on the basis of following points:

 Performance measure which defines the success

criterion.
 Agent prior knowledge of its environment.
 Best possible actions that an agent can perform.
 The sequence of percepts.

BY Abdela Aman. 09/30/2021

 Rationality:
20

 What is rational at any given time depends

on four things:
 The performance measure that defines

the criterion of success.

– The agent's prior knowledge of the
environment.
– The actions that the agent can perform.
– The agent's percept sequence to date

BY Abdela Aman. 09/30/2021

 Task /Nature of Environments
21
PEAS Representation
 PEAS is a type of model on which an AI agent works upon.
 When we define an AI agent or rational agent, then we can
group its properties under PEAS representation model.
 It is made up of four words:
 P: Performance measure
 E: Environment
 A: Actuators
 S: Sensors

 Here performance measure is the objective for the success of

an agent's behavior.

BY Abdela Aman. 09/30/2021

 PEAS for self-driving cars:
22

BY Abdela Aman. 09/30/2021

 PEAS for self-driving cars:
23

 Let's suppose a self-driving car then PEAS representation will be:

 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

BY Abdela Aman. 09/30/2021

 Example of Agents
24

BY Abdela Aman. 09/30/2021

 25
Types of AI Agents
 Agents can be grouped into five classes based on their degree of

perceived intelligence and capability.

 All these agents can improve their performance and generate better

action over the time.

 These are given below:

 Simple Reflex Agent

 Model-based reflex agent
 Goal-based agents
 Utility-based agent
 Learning agent
BY Abdela Aman. 09/30/2021
 1.Simple Reflex Agent
26

 The Simple reflex agents are the simplest agents.

 These agents take decisions on the basis of the current

percepts and ignore the rest of the percept history.

 These agents only succeed in the fully observable environment.

 The Simple reflex agent does not consider any part of percepts

history during their decision and action process.

 The Simple reflex agent works on Condition-action rule, which

means it maps the current state to action.

BY Abdela Aman. 09/30/2021

 Example
27

BY Abdela Aman. 09/30/2021

 Problem with it
28

 Problems for the simple reflex agent design approach:

 They have very limited intelligence

 They do not have knowledge of non-perceptual parts of the
current state
 Mostly too big to generate and to store.
 Not adaptive to changes in the environment.

BY Abdela Aman. 09/30/2021

 2. Model-based reflex agent
29
 The Model-based agent can work in a partially observable
environment, and track the situation.
 A model-based agent has two important factors:
 Model: It is knowledge about "how things happen in the world," so
it is called a Model-based agent.
 Internal State: It is a representation of the current state based
on percept history.
 These agents have the model, "which is knowledge of the world" and
based on the model they perform actions.
 Updating the agent state requires information about:
 How the world evolves

 How the agent's action affects the world.

BY Abdela Aman. 09/30/2021

 2. Model-based reflex agent
30

BY Abdela Aman. 09/30/2021

 3. Goal-based agents
31
 The knowledge of the current state environment is not always

sufficient to decide for an agent to what to do.

 The agent needs to know its goal which describes desirable

situations.
 Goal-based agents expand the capabilities of the model-based

agent by having the "goal" information.

 They choose an action, so that they can achieve the goal.

 These agents may have to consider a long sequence of possible

actions before deciding whether the goal is achieved or not.

BY Abdela Aman. 09/30/2021

 3. Goal-based agents
32

 The agent program can combine this with the model to

choose actions that achieve the goal.

 The knowledge that supports its decisions is represented

explicitly and can be modified, which makes these agents

more flexible.
 They usually require search and planning.

 The goal based agent’s behavior can easily be changed.

BY Abdela Aman. 09/30/2021

 3. Goal-based agents
33

BY Abdela Aman. 09/30/2021

 4. Utility-based agents
34

 These agents are similar to the goal-based agent but provide an extra

component of utility measurement which makes them different by

providing a measure of success at a given state.
 Utility-based agent act based not only goals but also the best way to

achieve the goal.

 The Utility-based agent is useful when there are multiple possible

alternatives, and an agent has to choose in order to perform the best

action.
 The utility function maps each state to a real number to check how

efficiently each action achieves the goals.

BY Abdela Aman. 09/30/2021

 4. Utility-based agents
35

 Goals alone are not enough to generate high-quality behavior

in most environments.
• For example, many action sequences will get the taxi to its
destination (thereby achieving the goal) but some are quicker,
safer, more reliable, or cheaper than others.
• Goals just provide a crude binary distinction between “happy”
and “unhappy” states.

BY Abdela Aman. 09/30/2021

 Utility-based agents
36

BY Abdela Aman. 09/30/2021

 5. Learning Agents
37
 A learning agent in AI is the type of agent which can learn from its past
experiences, or it has learning capabilities.
 It starts to act with basic knowledge and then able to act and adapt
automatically through learning.
 A learning agent has mainly four conceptual components, which are:
 Learning element: It is responsible for making improvements by
learning from environment
 Critic: Learning element takes feedback from critic which describes
that how well the agent is doing with respect to a fixed performance
standard.
 Performance element: It is responsible for selecting external
action
 Problem generator: This component is responsible for suggesting
actions that will lead to new and informative experiences.

BY Abdela Aman. 09/30/2021

 5. Learning Agents
38

BY Abdela Aman. 09/30/2021

 39

END!!!

BY Abdela Aman. 09/30/2021