That is a valid question. Why Neural Networks?
Neural Network is a fascinating technology, 50 years old,
but still not fully employed. And the question is why? Why
didn't Neural Network progress as fast as many other technologies?
Let us first take a look back …
The concept of neural networks has been around since the early
1950s, but was mostly dormant until the mid 1980s. One of
the first neural networks developed was the perceptron created
by a psychologist named Frank Rosenblatt in 1958. The perceptron
was a very simple system used to analyze data and visual patterns,
which generated a great deal of interest in AI community.
Unfortunately,
these earlier successes caused people to exaggerate the potential
of neural networks, particularly in light of the limitation in the
electronics then available.
Rosenblatt and other scientists claimed that eventually, with enough
complexity and speed, the perceptron would be able to solve almost
any problem.
In
1969, Marvin Minsky and Seymour Papert of MIT published an influential
book, which showed that the perceptron could never solve a class
of problems, and hinted at several other fundamental flaws in the
model.
Their
analysis combined with unfulfilled, outrageous claims convinced
the AI community; and the bodies that fund it; of the fruitlessness
of pursuing work with neural networks, and the majority of researchers
turned away from the approach.
The
result was to halt much of the funding and scientists working on
neural network type devices found it almost impossible to receive
funding.
This
period of stunted growth lasted through eighties where several events
caused a renewed interest. In 1982 John Hopfield of Caltech presented
a paper to the national Academy of Sciences. With clarity and mathematical
analysis, he showed how such networks could work and what they could
do.
By
1985 the American Institute of Physics began what has become an
annual meeting of Neural Networks for Computing. By 1987, the Institute
of Electrical and Electronic Engineer's (IEEE) first International
Conference on Neural Networks drew more than 1,800 attendees.
And
the 1990 US Department of Defense Small Business Innovation Research
Program named 16 topics, which specifically targeted neural networks.
By
then, the wheel turned again and growth started, but not with the
pace that one would wish to see. Over shadowed by Internet explosion,
processing limitations also contributed to the slow growth.
In
meantime Internet hype has settled down and processing power
is no showstopper anymore. Computerization of business and
personal transactions generate the flood of data that would
certainly contribute to machine learning and other modern
data analysis methods.
Thanks
to the availability of cheap microprocessors and recent discoveries
about DNA and human brain, artificial intelligence has gone from
being a fantasy to becoming a reality. In fact, most AI researchers
believe that it's only a matter of 20 to 30 years before machines
become at least as intelligent as humans.
Already
over 80% of Fortune 500 have Neural Net R&D programs and others
are realizing its importance.
Now
... Neural Network is back and this time ... to stay ...
Yet, its future, indeed the very key to the whole technology, lies
in commercial use.
Where is computer industry headed?
In 1982 IBM introduced the first PC with 64 KB Memory and 5 MB hard-drive
for about $3000. In 1991 the same money could buy you a PC with 16 MB
Memory and 1 GB hard-drive. In other words in just 15 years disk storage
has increased 200 times. A detail look at the trend indicates exponential
growth in both performance and storage space of computers.
That will obviously create demand for smarter and more flexible software
solutions. As example:
Businesses are already gathering gigabyte of data daily and that is
also rapidly growing. The old query and reporting tools are losing their
ability to keep up with amount of data and the information they can
provide. The newer technologies such as free-form query and OLAP are
certainly helping, but if you know what you are looking for. But what
about unknown patterns and variable dependencies buried in Terra Bytes
of data?
Yet everybody would agree that a natural direction for computers would
be Soft Computing and Artificial Intelligence. A path that will lead
us to solutions for our demanding future needs.
Trendy applications such as Data Mining, Business Intelligence and Robotics
have recognized that fact and are already utilizing AI in many different
ways. From Heuristic Algorithms, Fuzzy Logic, Genetic and Evolutionary
Algorithms to Neural Nets. Neural
Networks satisfies many requirements of futuristic applications such
as:
A
perfect match for Applications such as:
Intelligent Agents, Monitoring and Warning Systems, Advance Decision
Support System (ADSS), Process Automation, Intelligent Personal
Assistant and Smart devices.
But yet, Neural Nets are not without problems.
- The inner workings of neural networks are like "black boxes."
Some people have even called the use of neural networks "voodoo
engineering".
They learn and model based on experience, but they cannot explain and
justify their decisions (not that we human can!).
- Neural networks might have high accuracy, but not 100%, as we would
want. And unfortunately, not all applications can tolerate that level
of error.
But again, even if not exact, they still can be used in conjunction
with traditional methods to, for example, cutting down on time in search.
- Neural networks require lot of (sample) data for training purpose.
It may have been an obstacle few years ago, but as we move forward that
wouldn't be a problem. We are moving to an age where many of our activities
are digitally recorded. From bank transactions to phone conversations
and from medical history to lab results.
- As popularly portrayed, design and modeling of Neural Network is an
Art, rather than a science. But in contrast to other Arts, the outcome
of this one can be measured!
To
summarize, Neural Network is not the perfect solution for every
problem, but also there are lot more applications that can benefit
from ANN characteristics than we see today.
add
to favorites
