Saturday, June 29, 2013
ABSTRACT
Artificial intelligence
is combination of a computer science, psychology, philosophy, which deals with
different fields from machine vision to expert system. This paper gives
disciplined description about AI techniques, which explores knowledge and
further informs the history and process of development of AI including “HOLY
GRAIL” Turing’s test.
It enlightens various theories on artificial
neural networking, a mimic of human brain including parallel computation and
extends its emphasis on Expert systems, Common sense and Fuzzy logic, which are
the advanced developments of knowledge-based systems. Here, report exemplifies
various applications in fields such as Chess, Robotics, Defense and Revolution
in hearing care ( Adapto ™).
Further more we tried to focus
on various researches which are being carried out in different laboratories
around the world. This also debates on the existence of AI for future
generations. Finally, we have specified the consequences of AI with humanity
and ended with an interesting conclusion.
ARTIFICIAL INTELLIGENCE
INTRODUCTION
Artificial Intelligence (AI) is the
area of computer science focusing on creating machines that can engage on
behaviors that human considers intelligent. The dream of creating intelligent
machine has intrigued humans since ancient times and came into reality with the
advent of computers and 50 years of research. These smart machines can mimic
human thought, understand speech, beat the best human chess player and
countless other feats never before possible, equal or exceed human abilities,
and became an important part of most
business and Government operations as well as daily activities having the near
future with AI impacting human lives.
WHAT IS ‘AI’?
AI is a combination of computer
science, physiology, philosophy. AI is the study of the computations that make
the machine possible to perceive, reason and act. AI helps in designing smart
machines that can “think”.
In order to classify machines as thinking it
is necessary to define as it embodies all off the knowledge and feats, both
conscious and unconscious, which is acquired through study and experience,
highly refined sight and sound perception, thought, imagination; the ability to
converse, read, drive a car, memorize and recall facts, express and fell
emotions and much more just like remembering a face not seen for thirty or more
years, or to build and send rockets to the moon. It is those capabilities that
set homosapiens a part from other forms of living things. Hence, researches in
the fields to satisfy the conditions and requirements have aided scientists in
building intelligent machines which is one of the most challenging approaches
facing experts in building systems that mimic the behavior of human brain, made
up of billions of neurons and arguably the most complex matter is in the
universe.
HISTORY OF AI
Evidence
of artificial intelligence folklore can be traced back to ancient Egypt
NEWELL and SIMON developed the logic theory, considered by many to be the
first AI program representing each problem as a tree model. In 1956 John
McCarthy regarded as the father of AI organized a conference named “The
Dartmouth summer research project on artificial intelligence”, which brought
together the founders in AI and served to lay the ground work for the future of
AI research. From the researches made for the development of AI it is found
that neural networks in AI act as a mimic to that of the networking in human
brain.
NEURAL NETWORKING
INTRODUCTION
In the quest to create intelligent machines the
field of AI has split into several different approaches based on the opinions
about the most promising methods and theories with two approaches; bottom-up
and top-down. Bottom-up theorists believe the best way to achieve AI is to build
electronic replicas of the human brains complex network of resources, while the
top-down approaches attempts to mimic the brains behavior with computer
programs.
NEURAL NETWORKS AND PARALLEL
COMPUTATION
The
human brain is made up of a web of billions of cells called neurons and
understanding its complexities is seen as one of the last frontiers in
scientific research. It is the aim of AI research that prefers this bottom-up
approach the construct electronic circuits that act as neurons do in the human
brain. Although much of the working of the brain remains unknown, the complex
networks of neurons are what give human intelligent characteristics. The
average human brains weighs about 3.3 pounds and contains an estimated no 1012
neurons. The neurons and their inter connection capabilities provide about 1014
bits of potential storage capacity. But it self-neurons is not intelligent but
when grouped together neurons are able to pass electronic signals together
networks. The neurons ‘firing’ passes a signal to the next in the chain.
Research has shown that a signal received by a neuron travels through the
dendrite region and down the axon separating nerve cells is a gap called the
synapse in order for the signal to be transferred to the next neuron the signal
must be converted from electrical energy. The signal can then be received by
the next neuron and processed.
An
important back of mathematics logic binary numbers were also the basis of the
AI this is the basis of computer simulated neural networks also know as
parallel computing
McCulloch
and Pitts, using Boole’s principles wrote a paper on neural network theory. It
state that one of the level of single neuron the release of failure to release
an impulse was the basis by which the brain makes true/false decisions using
the idea of feedback theory. Their theory showed how the firing of signals
between connected neurons cause the brain to make decisions. McCulloch and
Pitts theory is the basis of the artificial neural network theory.
With
new top-down methods becoming popular parallel computing up put on hold. Now
rural network are making a return and same researches believe that with new
computer architectures parallel computing and the bottom theory will be a
during factor in creating AI. These theories touched on some of the main
methods used to create intelligence. These approaches have been applied to
different systems using variety of programs, which are based on knowledge.
KNOWLEDGE BASED SYSTEMS
Knowledge is defined as the body of facts and principles accumulated
by human kind or the act. Generally the limited usage of laws and axioms were less
effective in solving problems of any complexity. These realizations eventually
led to the design known as knowledge based systems.
Knowledge
based systems get their power from the expert knowledge that has been coded
into fact rules, heuristics and procedures.. The knowledge based systems were
real world problem solvers, tackling tasks such as determining complex chemical
structures of atomic constituents and mass spectra data from samples of the
compounds and later performing diagnosis of infectious blood diseases.
Further researches on the knowledge based systems inferences the
fields such as
1.
Experts systems
2.
Common sense
3.
Fuzzy logic
EXPERT SYSTEMS
Expert
systems, one of the developments of knowledge based systems; solve problems that
are normally solved by human experts which require a kind of intelligence and
reasoning mechanism. To solve expert level problems, expert systems need access
to a substantial domain knowledge base, which was built as efficiently as
possible.
The most widely used way of representing
knowledge in expert systems is as a set of production rules. For example, say
the situation was birthday party. A system could call on its birthday frame and
use the information contained in the frame to apply the situation. The system
knows that there is usually cake and presents because of the information
contained in the knowledge frame. The use of frames also allows the system to
add knowledge. Experts systems are even applicable in forecasting weather
Because
of the large storage capacity computers ranging in 1012, expert
systems had the potential to interpret statistics in order to solve the problem
like a detective solves a mystery. For example, charts like these represent the
logic of Expert systems.
Expert systems have the power
and range to aid to benefit and some cases replace humans and human experts, if
used with discretion will benefit human kind.
COMMON SENSE
Computers
have entirely deserved reputation for lacking common sense. A number of
techniques can be used to enable an AI program to represent and reason with
common sense knowledge. Non – atomic logic’s can support default reasoning, an
important aspect of common sense. As of yet, no program can match the common
sense reasoning powers. This is due, impart to the large amount of knowledge
required for common sense. Memory is another key aspect to common sense.
·
MEMORY ORGANIZATION Memory is central to common sense
behavior. Memory is also the basis for learning. A system that cannot learn
cannot, in practice, posse common sense. Psychology and AI seek to address
these issues. It is difficult to know which script is retrieve. It is hard to
modify a script. More recent work reduces script individual ‘scenes’, which can
be made shade across multiple structures.
·
CASE – BASED REASONING Computer systems that solve new problems by analogy with old ones
are often called as case – based reasoning (CBR). A CBR system draws its power
from a large case library, rather than from a set of basic principles. A
general CBR system must be able to learn a proper set of indices from
experience. The idea is to starts solving problems with a heuristic search
engine.
FUZZY LOGIC
Fuzzy
logic acts as an alternative for representing some kinds of uncertain
knowledge. It is possible to define a reasoning system based on techniques for
combining distributions. Such reasoning’s have been applied in control systems
as Simplified control of Robots
(a) Software design for industrial projects
(b) Prediction for early recognition of earth
quakes.
APPLICATIONS
The study of issue of AI for quit some time now made to
know all the terms and facts related to it. But what is really needed to know
is what can be done to get human hand on some AI today. AI with its learning
capabilities can accomplish these tasks. The advanced applications of AI are
Ø ROBOTICS The applications of automatic machines
to perform tasks are traditionally done by humans. Many are now capable of
simple decision making with out the intervention of operator using AI
techniques. Some of the latest and advanced robots that can be seen are : Frog
bout Robot, Jumping Robot
Ø CHESS AI
based game playing programs combine intelligence with entertainment. On game
with strong AI ties is chess. World chess champion playing programs can seen
ahead twenty plus moves in advance for each move they make. In three minutes,
Deep thought considers 126 million moves, while human chess master on average
considers less than 2 moves.
§ AI IN DEFENSE The
military put AI based hardware to the test of war during desert storm. AI based
technologies in missile systems, heads up displays and other advancements.
§ AI REACHING THE EAR According
to press reports on October 20th 2001 in
ADVANCE RESEARCHES
Many famous research institutes around the world are
undertaking the important advanced researches related to AI. Some of those
researches are:
q ROBOTICS Coco is
a small ape like robot, to explore humanoid intelligence. It requires large amount
of parallel computation in order to support rich real time and multi – model
sensory input and complex behavioral output.
Current Researches on
Ø Brain infra structure for coco
Ø A mobile coco for social interaction
Ø Medial axis representation for the
analysis of human shaping image.
Some other famous Robots are COG, KISMAT,
and MACACO.
q
MEDICAL VISION GROUP: Its goal is
to develop new algorithms for medical images analysis and visualization of
medical imaginary as well as to build vision based system for surgical
navigation and surgical planning. In this project “Image Guided System”, “Medical
Image Analysis” and “Visualization” are the main three categories.
q PROJECT ARIES: Today’s parallel machines suffer from
unvarying degrees from poor programmability, inadequate network performance and
limited scalability. The Aries research group is exploring novel processor,
network and software constructs to overcome these limitations.
CONCLUSION
In
the coming decades one should not expect that human race will become extinct
and be replaced by AI. One can expect that classical AI will go on producing
more and more sophisticated applications in restricted domains. But any time
one can expect common sense will continue to be disappointed may even in the
past. At vulnerable points these will continue to be exposed as “blind
automata”. In conclusion, we won’t see full AI in human lives, but should live
to get a good feel for whether are not it is possible and how it could be
achieved by our descendants. At last it can be said, “AI is possible…but it
won’t happen”.
|
