ARTIFICIAL

INTELLIGENCE(AI)
Chapter II
INTRODUCTION TO INTELLIGENT AGENTS
Chapter Outline

 Introduction to Intelligent Agents

 Introduction

 Agents and Environments

 Acting of Intelligent Agents (Rationality)
 Structure of Intelligent Agents
 Agent Types
 Important Concepts and Terms
Introduction

 AI has two major roles:

1. Study the intelligent part concerned with humans.
 It have two goals
 Scientific Goal:
 To determine knowledge representation, learning, rule
systems search, and so on
 Explain various sorts of real intelligence.
Introduction

 Engineering Goal:
 To solve real world problems using AI techniques such
as Knowledge representation, learning, rule systems,
search, and so on.
 Traditionally, computer scientists and engineers have
been more interested in the engineering goal, while
 Psychologists, philosophers and cognitive scientists
have been more interested in the scientific goal.
Introduction

2. Design those actions using computers

(Representation).
 In order to design intelligent systems, it is important to
categorize them into four categories (Luger and
Stubberfield 1993), (Russell and Norvig, 2003)
 1. Systems that think like humans
 2. Systems that think rationally
 3. Systems that behave like humans
 4. Systems that behave rationally
Introduction
Introduction

 Cognitive Science: Think Human-Like

 Requires a model for human cognition. Precise enough
models allow simulation by computers.
 Focus is not just on behavior and I/O, but looks like
reasoning process.
 Goal is not just to produce human-like behavior but to
produce a sequence of steps of the reasoning process,
similar to the steps followed by a human in solving the
same task.
Introduction

 Laws of thought: Think Rationally

 The study of mental faculties through the use of
computational models; that it is, the study of computations
that make it possible to perceive reason and act.
 Focus is on inference mechanisms that are probably correct
and guarantee an optimal solution.
 Goal is to formalize the reasoning process as a system of
logical rules and procedures of inference.
 Develop systems of representation to allow inferences to
be like “Socrates is a man. All men are mortal. Therefore
Socrates is mortal”.
Introduction

 Turing Test: Act Human-Like

 The art of creating machines that perform functions
requiring intelligence when performed by people; that it is
the study of, how to make computers do things which, at
the moment, people do better.
 Focus is on action, and not intelligent behavior centered
around the representation of the world.
Introduction

 Rational agent: Act Rationally

 Tries to explain and emulate intelligent behavior in terms of
computational process; that it is concerned with the
automation of the intelligence.
 Focus is on systems that act sufficiently if not optimally in
all situations.
 Goal is to develop systems that are rational and sufficient.
Agents and Environments

 An agent is an entity that perceives and acts.

 Abstractly, an agent is a function from percept
histories to actions:
[f: P*  A]
 For any given class of environments and tasks, we
seek the agent (or class of agents) with the best
performance.
Agents and Environments
Agents and Environments

 The agent able to sense the environment with the

help of sensors, the environment
percept/detected/observed by the agent through
sensors.
 Then the sensors sends the signal or energy to the
actuators.
 The actuators performs the action means they
converts this signal or energy into motions.
 So the action is seen on the environment.
Agents and Environments
AI Agent Terminologies

 An agent:
 Is anything that can perceive/observe/detect its
environment through sensors and acts upon that
environment through actuators.
 Agents can be three types:
 A human agent:
 Has sensory organs such as eyes, ears, nose, tongue and
skin and has actuators such as hands, legs, mouth.
AI Agent Terminologies

 A robotic agent:
 Has cameras and infrared range finders for the sensors,
and various motors/tires used as actuators.
 A software agent:
 Has encoded bit strings as its programs and actions (by key
strokes).
AI Agent Terminologies

 Sensor:
 It is a device which perceives/observes/detects the change
in the environment and sends the information to the other
electronic devices/actuators/effectors.
 Actuators/effectors:
 It is component of machine that converts energy into
motion.
 It is responsible for moving and controlling of the system.
 It is the device which affects the environment.
 It can be legs, wheels, arms, fingers and display screens.
AI Agent Terminologies

 Performance Measure of Agent

 It is the criteria, which determines how successful an agent is.
 Behavior of Agent
 It is the action that agent performs after any given sequence of percepts.
 Percept
 It is agent’s perceptual inputs at a given instance.
 Percept Sequence
 It is the history of all that an agent has perceived till date.
 Agent Function
 It is a map from the precept sequence to an action.
Agent Rationality

 Rationality is nothing but status of being

reasonable, sensible, and having good sense of
judgment.
 Rationality is concerned with expected actions and
results depending upon what the agent has
perceived.
 Performing actions with the aim of obtaining useful
information is an important part of rationality.
Ideal Rational Agent

 It is which has a capable of doing expected actions to

maximize its performance measure, on the basis of:
 Its percept sequence
 Its built-in knowledge base
 They always performs right action, where the right action
means the action that causes the agent to be most successful
in the given percept sequence.
 The problem the agent solves is characterized by Performance
Measure, Environment, Actuators, and Sensors (PEAS).
Rationality of an agent depends

 The performance measures, which determine the

degree of success.
 Agent’s Percept Sequence till now.
 The agent’s prior knowledge about the
environment.
 The actions that the agent can carry out.
 Structure of Intelligent Agents.
Agent’s structure can be viewed as:

 Agent = Architecture + Agent Program

 Architecture = the machinery that an agent
executes on.
 Agent Program = an implementation of an agent
function.
Types of Agents

 Agents can be grouped into four classes based on

their degree of perceived intelligence and capability:
 Simple Reflex Agents: Complete Observation of
only current state
 Model-Based Reflex Agents: partially
observable, depends on the percept history
 Goal-Based Agents: selecting the one which
reaches a goal state
 Utility-Based Agents: it bother about happiness
Types of Agents

 Simple reflex agents

 It works only on the basis of current perception and it does not
bother about the previous state in which the system was.
 It ignore the rest of the percept history and act only on the basis
of the current percept.
 Percept history is the history of all that an agent has perceived
till date.
 It is based on the condition-action rule
 A condition-action rule is a rule that maps a state i.e., condition
to an action.
 If the condition is true, then action is taken, else not taken.
Types of Agents

 Simple reflex agents

Types of Agents

 Problems with simple reflex agents are:

 They choose actions only based on the current percept.
 They are rational only if a correct decision is made only on
the basis of current precept.
 Their environment is completely observable.
 Very limited intelligence.
 No knowledge of non-perceptual parts of state.  Usually
too big to generate and store.
Types of Agents

 Model Based Reflex Agent

 It works by finding a rule whose condition matches the current
situation.
 It can handle partially observable environments by use of
model about the world.
 The agent has to keep track of internal state which is adjusted
by each percept and that depends on the percept history.
 Updating the state requires the information about:
 How the world evolves in-dependently from the agent, and
 How the agent actions affects the world.
Types of Agents

 Model Based Reflex Agent

Types of Agents

 Goal Based Agent

 Agents take decision based on how far they are currently from
their goal(description of desirable situations).
 Their every action is intended to reduce its distance from goal.
 This allows the agent a way to choose among multiple
possibilities, selecting the one which reaches a goal state.
 The knowledge that supports its decisions is represented explicitly
and can be modified, which makes these agents more flexible.
 Usually require search and planning.
 The goal based agent’s behavior can easily be changed.
Types of Agents

 Goal Based Agent

Types of Agents

 Utility Based Agents

 The agents which are developed having their end uses as
building blocks are called utility based agents.
 When there are multiple possible alternatives, then to
decide which one is best, utility based agents are used.
 They choose actions based on a preference(utility) for each
state. Sometimes achieving the desired goal is not enough.
 We may look for quicker, safer, cheaper trip to reach a
destination.
Types of Agents

 Utility Based Agents

 Agent happiness should be taken into consideration.
 Utility describes how “happy” the agent is.
 Because of the uncertainty in the world, a utility agent
chooses the action that maximizes the expected utility.
 A utility function maps a state onto a real number which
describes the associated degree of happiness.
Types of Agents

 Utility Based Agents

Interacting with the Environment

 In order to enable intelligent behaviour, we will have

to interact with our environment.
 Properly intelligent systems may be expected to:
 Sensory input
Vision, Sound, …
 Interact with humans
Understand language, recognise speech,
generate text, speech and graphics, …
 Modify the environment
Robotics
Properties of an Environment

 The environment has multifold properties:

 Discrete / Continuous
 If there are a limited number of distinct, clearly defined, states
of the environment, the environment is discrete (For example,
chess); otherwise it is continuous (For example, driving).
 Complete Observable / Partially Observable
 If it is possible to determine the complete state of the
environment at each time point from the percepts it is
observable; otherwise it is only partially observable.
Properties of an Environment

 Static / Dynamic
 If the environment does not change while an agent is
acting, then it is static; otherwise it is dynamic.
 Single agent / Multiple agents
 The environment may contain other agents which may be
of the same or different kind as that of the agent.
 Accessible / Inaccessible
 If the agent’s sensory apparatus can have access to the
complete state of the environment, then the environment
is accessible to that agent.
Properties of an Environment

 Deterministic / Non-deterministic
 If the next state of the environment is completely
determined by the current state and the actions of the
agent, then the environment is deterministic;
otherwise it is non-deterministic.
The Disadvantages of AI

 Increased costs
 Difficulty
with software development - slow and
expensive
 Few experienced programmers
 Few practical products have reached the market
as yet.
  Chapter III Outline
 Problem Solving by
Searching
End of Chapter  Problem Solving Agents

II 


Problem Formulation
Search Strategies
Next Chapter III  Constraint Satisfaction
Search
 Games as Search
Problems