UrbanPro
true

Learn IT Automation from the Best Tutors

  • Affordable fees
  • 1-1 or Group class
  • Flexible Timings
  • Verified Tutors

Search in

Book On Artificial Intelligence 1.1 Intelligent Agents

Sivakumar C
23/08/2017 0 0

Forward

 

This book is forwarded to the ambitious engineering students and staffs who are guiding them in the winning direction.

 

Abstract

This book will make you understand the basics of Artificial Intelligence (AI from herein after), how to make the machines to think (though it is not yet possible), Game planning, Real world applications and related topics.

           

As a pre requisite, we would expect you to be fluent in the language (English) and we promise we will make a Tamil, Hindi, Telugu, Malayalam, Kannada, Marathi and other regional languages translation of this book available As Soon As Possible (ASAP).

 

Apart from that, we wish you happy learning. Best Wishes. – Author Nick Name: Stella Peter Real Name: C.Sivakumar, B.E., (M.Tech)., PGDNLP.,

You may reach the author by email: futechteam@gmail.com. Phone number:+91 9843492622 and address: 4 Middle St 31Z 5D North St Aranarai Colony Perambalur-621212 TamilNadu India

 

Unit 1: Introduction

Definition of AI

            AI, Artificial Intelligence is the ability to make an artifact system to perceive reason, learn and act with in the legal boundaries and socially acceptable standards.

            Artificial Intelligence is the ability to make computers to PERCIEVE, REASON & ACT Rationally. To make the computer system to THINK & ACT like humans, to include minds into the machines.

 

            An Agent (or a ROBOT) PERCEIVES (reads or senses) its environment using SENSORS(Detectors) and acts upon that environment through Actuators(external limbs).

           

Future challenges of AI:

            It is to to make the machines to think.

As Alan Turing said in his essay on “Computing Machines and Intelligence”, “there is only short distance ahead but a LOT TO BE DONE

 

1.1 Intelligent Agents

            The nature and characteristics of agents are discussed here. The Agent is expected to be intelligent, perfect or at least near perfect, able to tackle diversified and varied environments, new surroundings (LEARNING AGENTS that update their knowledge base or data base). According to this the resulting agent types will vary.

  1. The heart of Artificial Intelligence is the rational agents or rationality
  2. Rationality is applied to many types of agents working in complex and varied environments like Road (Taxi Agent), Air ( Auto Pilot Agent), Class Room (Lecturer or Teacher Agent)
  3. In order to be intelligent, the agents need to be coupled to the environments they are expected to perform in.
  4. An agent is rational if it behaves better than another agent in the defined capacity.
  5. The behavior of an agent depends on the environment because the level of difficulty varies from one environment to another – for example a chess playing agent deals with a static environment of a chess board or a computer monitor whereas a foot ball playing agent have to deal with a more complex and difficult environment of the huge foot ball ground and also it is dynamic.

 

 

Human

Rational

 

THINK

Think Like Human Cognitive Modeling

Think Rationally Rules and regulations of thought patterns

 

ACT

Act Like Human Turing Test

Act Rationally Rational agent versus irrational agent

 

 

 

 

 

 

Intelligent Agent is a System that PERCEIVES (Receives Input from Sensors like Camera for Seeing, looking or Eyeing, Mike for hearing or Earing and Sends OUTPUT through Actuators like Wheels for Moving or Walking or Legging, Speakers for Speaking)        An Agent (or a ROBOT) PERCEIVES (reads or senses) its environment using SENSORS (Detectors) and acts upon that environment through Actuators (external limbs).

            Artificial Intelligence is the ability to make computers to PERCIEVE, REASON & ACT Rationally. To make the computer system to THINK & ACT like humans, to include minds into the machines.

 

Acting Humanly: Preferably favorably

 

Examples:

 

Robotic Agent: Sensors: Cameras(for eyeing or seeing), IR, mike, Actuators, Motors, Speakers

 

Software agent(MS Excell): Sensory Inputs: Keyboard strokes fills cells

Actuator Outputs: Display formula results on screen

 

Inputs are percepts; Percept Sequence is the history of all inputs perceived(received) by an agent.

 

Q: Differentiate between agent function & agent program

Answer:

Agent Program

Agent Function

Internal implementation of an agent function(Operation)

External characterization of an agent in the form a table

Concrete implementation

Abstract mathematical description

Runs on agent architecture

Runs on the agent's external world

What the agent thinks

What the agent does (acts)

 

Example: Vacuum cleaner world:

Locations: Square A, Square B

 

Vacuum agent's Percepts: 1: Which square it is in; 2: whether my square is dirty?

 

Vacuum agent's Actions: 1: MOVE left, 2: MOVE right, 3: remove dirt, 4: DO nothing

 

 

 

Human

Rational

 

THINK

Think Like Human Cognitive Modelling

Think Rationally Rules and regulations of thought patterns

 

ACT

Act Like Human Turing Test

Act Rationally Rational agent versus irrational agent

 

 

 

 

 

 

AI was founded on Philosophy when Socrates asked how you can differentiate between pious and impious. Mathematics and economics added significantly. Neuroscience led to research studies on neural networks and brain mappings. Psychology deserves a chapter on its own. Other foundations of AI includes Philosophy, Mathematics, Economics, Neuroscience, Psychology, computer science & engineering, control theory, cybernatics, logistics & linguistics, first order logic, propositional logic.

 

Computer engineering paralleled with Software engineering includes standard concepts like SDLC Software Development Life Cycle where it flows through the following stages:

SDLC Stage

Deliverable

 

Requirement Analysis

Feasibility Report

 

System Analysis

 

 

Cost Analysis

Cost Estimate Report

 

Time Analysis

Shedule

 

Detailed Data Base design

Data Flow Diagram

 

Program Design

Coding (Source Code)

 

Testing

Test Cases

 

User Acceptance Testing

Sign off

 

Unit Test

Unit Test cases

 

String Test

String Test cases

 

End to End Test

ETE Test cases

 

 

 

Major Milestones in the history of Artificial Intelligence:

            Starting with some early gestations between 1943 to 1955, AI was born in 1956. There were motivations and some expectation between 1952 and 1969.  But AI became real between 1966 and 1973. The seed to make AI an industry was founded in between 1969 to 1979 by knowledge based systems. The brain waved neural networks came back in 1986 and they are still alive now as well. AI started picking up more scientific methods in 1987 and it continues to do so until today. The concept of agents(widely used in the science fiction movie Matrix) was introduces in the year 1995 and people are fond of them until present. Very large data sets like data base Terra data became reality in the year 2001 and the need for huge data bases keeps growing day by day.

 

 

 

 

 

The above is a humanoid Robot called Topio, who managed to play TableTennis in the international Robot exihibition(IREX) in Tokyo, Japan (year 2009)

 

 

Turing Test:

            Introduced by the Alan Mr.Turing (1912-1954), a mathematics graduate from Kings College, Cambridge.

            A system passes the Turing test if the interrogator cannot differentiate between the answers coming from the human answerer and the system answerer. Which means the system had managed to simulate (mimic) the human behavior perfectly including emotions like crying and laughing.

 

 
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The above diagram depicts a typical Turing test where the Interrogator Lady receives answers from both the human answerer and the computer system answerer. The system will pass the test if she (the interrogator) cannot differentiate between the set of answers coming from the human and the computer system.

            A simple example can be the Tamil Movie ENTHIRAN by Dr Vasi (Mr. Rajnikanth) and Ms. Sana and Chitty the Robot (well humanoid in fact) .  For those who have not seen the movie, here is a brief: Doctor Vasi designs and the humanoid Chitty as a mirror image of him. Hence Chitty and Vasi looks the same. Sana is the partner and would be wife of Dr Vasi. Now as per the Turing test, even if Sana can not differentiate between Vasi and Chitty then Chitty passes the Turing test.

A computer needs the following abilities in order to pass the Turing test:

  1. NLP Natural Language Processing: T0 effectively communicate in English(Nasal, verbal stops, Parts of speech, vowels, consonants, Approximants, Trills, Flaps, Laterals etc)

Speech sound classification

 

Speech sound classification is classified according to the state and direction of air stream and degree of air passage. (Push air stream outside: explosives; push air stream inside: implosives.) When the air stream is pushed outside then those sounds are called explosives. When the air stream is pushed inside then those sounds are called implosives.

 

            Speech sounds are further classified as per degree of stricture, plosives, and affricates, fricatives nasals laterals rolls (trills), frictionless continuants and vowels.

 

            Distinction based on manner of articulation. * The way you pronounce.

 

                        Active articulators: lips tongue parts, vocal cards

 

Passive articulators:  upper front teeth alveolar ridge are parts of roof of mouth (hard, soft plate).

 

                        Bilabials, labiodentals alveolar palatal, velar glottal distinction based on point of articulation which active and which passive articulators are involved

 

                        Vowels: acoustic characterizati

0 Dislike
Follow 1

Please Enter a comment

Submit

Other Lessons for You

TestNG Annotations and its sequence
public class TestNGAnnotations { @BeforeMethod public void beforeM() { System.out.println("Before Method"); } @AfterMethod public void afterMethod() { System.out.println("After Method"); } @BeforeClass...
S

Sarthak C.

0 0
0

Big Data Hadoop
- Big data is a complex term which is difficult to manage a traditional system like RDBMS, DB2, SQL etc. - Hadoop is an open-source software framework, for storing massive data and running applications...

Selenium waits
In Selenium web driver, the biggest challenge is to synchronize your script with the Web application under test. To make this synchronization work, we have concept of waits. Although we can use Thread.sleep()...
X

Looking for IT Automation Classes?

The best tutors for IT Automation Classes are on UrbanPro

  • Select the best Tutor
  • Book & Attend a Free Demo
  • Pay and start Learning

Learn IT Automation with the Best Tutors

The best Tutors for IT Automation Classes are on UrbanPro

This website uses cookies

We use cookies to improve user experience. Choose what cookies you allow us to use. You can read more about our Cookie Policy in our Privacy Policy

Accept All
Decline All

UrbanPro.com is India's largest network of most trusted tutors and institutes. Over 55 lakh students rely on UrbanPro.com, to fulfill their learning requirements across 1,000+ categories. Using UrbanPro.com, parents, and students can compare multiple Tutors and Institutes and choose the one that best suits their requirements. More than 7.5 lakh verified Tutors and Institutes are helping millions of students every day and growing their tutoring business on UrbanPro.com. Whether you are looking for a tutor to learn mathematics, a German language trainer to brush up your German language skills or an institute to upgrade your IT skills, we have got the best selection of Tutors and Training Institutes for you. Read more