Scribd
Upload
11 views19 pages

AI Intro 1final

AIML introduction

Uploaded by

iam shrood
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
11 views19 pages

AI Intro 1final

AIML introduction

Uploaded by

iam shrood
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/ 19

Artificial Intelligence

It is a study of how to make computers “do things”


that are at a moment done better by people.

AI is Artificial Intelligence till it is achieved; After


which AI reduces to Already Implemented.
Intelligence

“The capacity to learn and solve problems”

in particular,

1. the ability to solve novel problems


2. the ability to act rationally
3. the ability to act like humans.

Artificial Intelligence

1. Build and understand intelligent entities or agents


2. Main approaches: “engineering” versus “cognitive
modeling”
AI Problems

Game Playing
Theorem Proving Both need above average intelligence

Every day tasks -- > Common sense reasoning  General Problem Solving (GPS)

No need for large knowledge base about a particular problem domain. Only
simple tasks are chosen

Example : Pick and place, Track a path, Move in a given direction etc.
AI Problems

Ability to handle large Knowledge base


Having done basis tasks, new tasks could be attempted

Perception -- Vision and Speech


Natural language understanding
Problem solving in specialized domains
Medical diagnosis or Chemical analysis

Perception : Knowing the world around -- > Crucial to survival


Perceptual task is difficult --? It involves analog data rather than digital data

Understanding language is an AI task so also is translation


.
AI Problems
More Specialized tasks
Requires high intelligence -> Engineering Design, Scientific discovery, Financial
analysis and planning
AI Problems - Broad categories
Vision, Speech
Mundane Tasks
Understanding and
Translation, Reasoning,
Robot control

Games – Chess, Checkers


Formal Tasks
Maths – Geometry, Logic
Proving Theorems,
Integral calculus

Expert Tasks Engineering – Design,


Fault finding, Planning
Scientific analysis
Financial Analysis
AI Scope and Future
AI is flourishing as a practical discipline and not as
a pure research
Expert Systems various engineering domains provide part or full
solutions to all practical problems that previously required human
expertise(Which is scarse)
Underlying Assumptions

Physical Symbol Hypothesis

• e.g. Turing machine model of information-


processing
• associated with classical, symbolic AI

Symbols are physical patterns


Symbols can be combined to form complex symbol structures

The system contains processes for manipulating complex symbol


structures
The processes for representing complex symbol structures can
themselves by symbolically represented within the system
Underlying Assumptions

Physical Symbol System Hypothesis(PSSH)

Basic components of a representation

• description of given situation


• operators for changing the situation
• a goal situation
• tests to determine whether the goal has been
reached
Underlying Assumptions

Examples of physical symbol systems include:

Formal logic: the symbols are words like "and", "or", "not", "for all x" and so on. The
expressions are statements in formal logic which can be true or false. The processes are
the rules of logical deduction.
Algebra: the symbols are "+", "×", "x", "y", "1", "2", "3", etc. The expressions are
equations. The processes are the rules of algebra, that allow one to manipulate a
mathematical expression and retain its truth.
A digital computer: the symbols are zeros and ones of computer memory, the processes
are the operations of the CPU that change memory.
Chess: the symbols are the pieces, the processes are the legal chess moves, the
expressions are the positions of all the pieces on the board.
The physical symbol system hypothesis claims that both of these are also examples of
physical symbol systems:
Intelligent human thought: the symbols are encoded in our brains. The expressions are
thoughts. The processes are the mental operations of thinking.
A running artificial intelligence program: The symbols are data. The expressions are more
data. The processes are programs that manipulate the data.
Underlying Assumptions

Physical Symbol System Hypothesis(PSSH)

Problem space = branching tree of achievable situations


defined by potential application of operators to initial situation
[e.g. chess]

Physical Symbol System Hypothesis : A physical system has necessary


and sufficient means for general intelligent action.
Ex. Symbol  Pawn on chess board Action -> Legal move
AI Technique

AI problems span a broad spectrum.

“Are there any techniques that are appropriate for solution of a


variety of these Problems?”

Ans : Yes…

All AI techniques require knowledge to act on symbols…

Properties of knowledge : It is voluminous, Hard to characterize,


constantly changing, Differs from data being organized that it is used.
Definition of AI Technique

“AI technique is a method that exploits knowledge and should be


represented such that”

1. Knowledge captures generalizations.


2. Is to be understood by people who use it.
3. It can be easily modified to correct errors and reflect real world
4. It can be used in great many situations
5. It can be used to help overcome its own sheer bulk by
narrowing the range of possibilites.
Examples :

Tick Tack Toe…


Level of the Model
Question :

“What is our goal in trying to produce programs that do the intelligent


things that people do?”

Goals :

1. Produce programs that do tasks the same way people do..

2. Produce programs that simply do the tasks whatever way


appears easiest.
Level of the Model

Goal 1 Model

Produce programs that do tasks the same way people do..

Programs that do not fit our Programs that Attempt to model


definition of AI task human performance and fit our
definition of AI task
Problems the computer can Problems that are non-trivial for a
easily solve : Memorizing computer : Read news paper story
syllables and answer a question
Level of the Model
McClloch & Pitts Model

Based on Biological Neuron called perceptrons, but this


failed to produce minimal intelligent behavior. Newer Neural
network architectures were proposed that have overcome
limitations of perceptrons. All neurons in the NN work
parallelly as do a human brain works.
Criteria for Success

Turing Test

Room 1 Room 2

B
One Final word and Beyond

We must do our best to develop programs to solve AI tasks to help understand how
human beings think and act..

All AI techniques must contain a lot of knowledge..if it has to solve some non-trivial
problems

But as size of knowledge grows, it becomes harder to access appropriate things


when needed.

You might also like