Download INTRODUCTION

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Incomplete Nature wikipedia , lookup

Human–computer interaction wikipedia , lookup

Wizard of Oz experiment wikipedia , lookup

Human-Computer Interaction Institute wikipedia , lookup

Artificial intelligence in video games wikipedia , lookup

Computer vision wikipedia , lookup

Technological singularity wikipedia , lookup

Expert system wikipedia , lookup

Embodied cognitive science wikipedia , lookup

Computer Go wikipedia , lookup

AI winter wikipedia , lookup

Intelligence explosion wikipedia , lookup

Knowledge representation and reasoning wikipedia , lookup

Existential risk from artificial general intelligence wikipedia , lookup

Philosophy of artificial intelligence wikipedia , lookup

Ethics of artificial intelligence wikipedia , lookup

History of artificial intelligence wikipedia , lookup

Transcript
ARTIFICIAL INTELLIGENCE
ARTIFICIAL INTELLIGENCE
Structure and Strategies for Complex Problem Solving
Author:
George F Luger and William Stebblfield
Edition:
Third
Publisher:
Addison Wisely
INTRODUCTION
 Artificial Intelligence (AI) is the design and study of
computer programs that behave intelligently.
 These programs are constructed to perform as humans
or as animals whose behavior we consider intelligent.
 AI researchers have written programs that control
nuclear power plants and diagnose problems in
complicated electronic devices.
 AI is concerned with programs that respond flexibly in
situations that were not specifically anticipated by the
programmer. A house cleaning robot should distinguish
between a scrap of tin foil and a diamond ring. A face
recognition system be able to identify the same face
with a different hat or hair cut.
Intelligence …
• Intelligence is defined as:
– The faculty of understanding;
– Capacity for learning, reasoning, and
understanding;
– Aptitude in grasping truths, relationships, facts,
meaning, etc.
– Mental alertness or quickness of understanding;
and
– Manifestation of a high mental capacity.
Artificial Intelligence ...
• The most common definition of Artificial
Intelligence is:
– The study of how to make computers do things
which, at the moment, people do better (Rich,
1991).
• Another way of defining AI is:
– The area of computer science focusing on creating
machines that can engage on behaviour that
humans consider intelligent.
Definition ...
• AI is a discipline that allows us to build
intelligent computers
(Systems/applications) which are capable of
learning, understanding, and developing a
sense to forecast, foretell, and foresee the
behaviour (Ali 1987).
Branches of AI …
Heuristics
Ontology
Pattern
Recognition
Expert
Systems
Representation
AI
Logical AI
Inference
Search
Learning from
Experience
Planning
 Much of the early work in AI focused on formal tasks,
such as game playing and theorem proving.
 Game playing and theorem proving share the property
that people who do them well are considered to be
displaying intelligence.
 AI is the part of computer science concerned with
designing intelligent computer systems, that is, computer
systems that exhibit the characteristics we associate with
intelligence in human behaviour - understanding language,
learning, reasoning and solving problems.
 The “Logic Theorist” was an early attempt to prove
mathematical theorems. It was able to prove several
theorems from the first chapter of a famous book of
Whitehead and Russell “Principia Mathematica”.
 Another mathematician Gelernter explored another area
of mathematics for writing a theorem proving program
e.g. geometry.
 Newell and Simon explored yet another area called
“common sense reasoning”. This area covers: Problem solving that we do every day when we decide
how to get to work in the morning.
 As well as actions and their consequences.
 To investigate this sort of reasoning they wrote a
program called “General Problem Solver”(GPS).
 AI research progressed and techniques for handling
larger amounts of world knowledge were developed.
Work also focused on following areas:-
 Perception (vision and speech).
 Natural Language Understanding.
 Problem solving in medical diagnosis
 Problem solving in chemical analysis.
 Engineering Design
 Scientific Discovery
 Financial Planning
 The problem areas where AI is now flourishing most as
a practical discipline are primarily the domains that
require only specialized expertise without the
assistance of commonsense knowledge. These are
called “ Expert Systems”.
 There are now thousands of Expert Systems in day-today operation throughout all areas of industry and
government around the world.
 Each of these systems attempts to solve part of or all of
a practical significant problem that previously required
human expertise.
A FEW DOMAINS OF AI
Mundane Tasks
 Perception
 Vision
 Speech
 Natural Language
 Understanding
 Generation
 Translation
 Common sense Reasoning
 Robot Control
A FEW DOMAINS OF AI
Formal Tasks
 Games
Tic-Tac-Toe
Checkers
Chess
Other Games
 Mathematics
Geometry
Logic
Integrated Calculus
Theorem Proving
A FEW DOMAINS OF AI
Expert Tasks
 Engineering
 Design
 Fault finding
 Manufacturing planning
 Scientific analysis
 Medical diagnosis
 Financial analysis
Knowledge
 One of the few hard and fast results to come out of
the first three decades of AI research is that
intelligence requires knowledge. There are a few
undesirable properties of knowledge: It is voluminous.
 It is hard to characterize accurately.
 It is constantly changing.
 It differs from data by being organized in a way
that corresponds to the ways it will be used.
AI Technique
 AI Technique is a method that exploits knowledge that
should be represented in such a way that: The knowledge captures generalizations, i.e. it is not necessary to
represent separately each individual situation. Instead situations
that share important properties are grouped together. If knowledge
does not have this property, inordinate amounts of memory and
updates will be required. So we usually call something without this
property “Data” rather than knowledge.
 In many AI domains, most of the “knowledge” a program has must
ultimately be provided by people in terms they understand.
Though in few AI applications, the “data” can be provided
automatically by taking readings from a variety of instruments.
AI Technique
 AI Technique is a method that exploits knowledge that
should be represented in such a way that: It can be easily modified to correct errors and to reflect changes in
the world and in our world view.
 It can be used in great many situations even if it is not totally
accurate or complete.
 It can be used to help overcome its own sheer bulk by helping to
narrow the range of possibilities that must usually be considered.
Artificial Intelligence - Theory
 AI is more than an engineering discipline, it is also a
subject of scientific investigation.
 Researchers construct theories about what AI programs
are capable of and test them with mathematical analysis
or
experiments.
 Theories are subjected to examination analytically by
developing mathematical abstractions and proving
theorems.
 They are also studied empirically by developing programs,
running experiments, and analyzing the results.
 The behavior of complex AI systems is difficult to predict.
Often researchers are surprised by the behavior of the
system that they build.
Examples of Artificial Intelligence - Theory
Inferring Structure from Motion in Machine Vision.
Machine vision is concerned with interpreting the information
contain in electronic camera images. Research has proved
that this information can be used to answer questions about
the structure and motion of the object captured in those
images.
Finding Consistent Hypothesis in Learning. In concept
learning, a system is given a set of examples of a target
concept and asked to find a hypothesis describing the
concept that is consistent with examples it has seen so far.
Probabilistic Inference in Diagnostic Reasoning. In
medical diagnosis, networks involving probabilities are used
to infer the most likely disease from a patient’s symptoms.
Examples of Artificial Intelligence - Theory
Search in Automated Planning. In planning for robots or
factories, it is desirable that an algorithm never consider the
same plan twice and that the algorithm always find a
solution to a planning problem if one exists.
Parsing Sentences in Language Understanding. Parsing
reveals the structure of sentences and is an important step
in automated language understanding. It is difficult to
program a computer to understand context of a spoken or
written sentence. Therefore, for context free languages,
parsing sentences is computationally easy. Though most
human languages are not context free languages.
Artificial Intelligence in Practice
 AI systems serve a wide variety of practical purposes.
There are programs that generate investment strategies
by predicting trends in the stock market, diagnose
patient illnesses suggesting treatment, and control
assembly robots in factories.
 People who work in AI consider themselves to be
engineers. They build practical tools. These tools called
AI systems are used to plan routes for airlines, build
cars in factories, and also play master level chess.
Examples of Artificial Intelligence Systems

Game Playing

Automated Reasoning and Theorem Proving

Expert Systems

Natural Language Understanding and Semantic
Modeling

Modeling Human Performance

Planning and Robotics

Languages and Environments for AI

Machine Learning
• Alan Turing, a pioneer in the
theory of computation, proposed
an intelligence test for computer
programs.
• A human judge is allowed to
interrogate a program through a
video terminal and a human
simultaneously.
• If the program can fool the judge into
believing that it is another human
responding rather than a computer,
then the program is judged to be
intelligent.
• One can imagine variants of this test
in which:– One manipulates a robot’s
environment to see how the robot
responds and judge the robot as
intelligent or not in comparison
with the responses of a human
worker under similar conditions.
Turing Test
The Measurement of
Intelligence of
AI Systems
Artificial Intelligence - Summary
It has been attempted to define artificial intelligence
through discussion of its major areas of research and
application. The primary concern of AI is to find effective
way to understand and apply intelligent problem solving,
planning and communication skills to wide range of
practical problems. In spite of the variety of problems
addressed in Artificial Intelligence research, a number of
important features emerge that seem common to all
divisions of the field, these are summarized as under:
The use of computers to do reasoning, pattern recognition, learning,
or some other form of inference.

A focus on problems that do not respond to algorithmic solutions.
This underlies the reliance on heuristic search as an AI problemsolving technique.
Artificial Intelligence - Summary

A concern with problem solving using inexact, missing, or poorly defined
information and the use of representational formalisms that enable the
programmer to compensate for these problems.

Reasoning about the significant qualitative features of a situation.

An attempt to deal with issues of semantic meaning as well as syntactic
form.

Answers that are neither exact nor optimal, but are in some sense
“sufficient.” This is a result of the essential reliance on heuristic problemsolving methods in situations where optimal or exact results are either
too expensive or not possible.

The use of large amounts of domain-specific knowledge in solving
problems. This is the basis of expert systems.
Assignment
Get the code of any games
Write down the worst effect
of AI on our society