Introduction to Intelligence

Systems
by
Mr. Mustafa Haitham Mohammed
Outline
• The concept of Machine Learning.
• Definition of intelligent systems.
– Difference from ordinary systems.
• Historical overview.
• Intelligent systems & Artificial Intelligence.
• Intelligent systems applications and uses.
• Main forms of intelligent systems.
Machine learning
• machine learning involves adaptive mechanisms that
enable computers to learn from experience, learn by
example and learn by analogy.
• Learning capabilities can improve the performance of
an intelligent system overtime.
• Such category of algorithms allows software
applications to become more accurate in predicting
outcomes without being explicitly reprogrammed.
– To build algorithms that can receive input data and
use statistical analysis to predict an output while updating
outputs as new data becomes available.
Machine Learning & Deep Learning
• Machine learning algorithms are often categorized
as supervised or unsupervised.
• Supervised algorithms require a data scientist or
analyst with machine learning skills to provide both
input and desired output.
• Unsupervised algorithms do not need to be trained
with desired outcome data.
– Instead, they use an iterative approach called deep
learning to review data and arrive at conclusions.
– Used for more complex processing tasks than supervised
learning systems.
• image recognition, speech-to-text, natural language generation
etc…
Intelligent Systems
• Intelligence dictionary definition:
1. Someone’s intelligence is their ability to understand
and learn things.
2. Intelligence is the ability to think and understand
instead of doing things by instinct or automatically.
• In order to think, a brain or an organ that resembles
the brain must exist:
1. To learn and understand things.
2. Problem solving and to decision making.
Cont’d
• Intelligent systems has the ability to evolve, self-
develop, self-learn continuously in order to reflect
the dynamically evolving environment.
• A truly intelligent system adapts itself to deal with
changes in problems (automatic learning).
• Few machines can do that at present.
• An ordinary system will simply repeat performing the
same steps in order to achieve a solution.
– No adaptation, donkey-work mechanism.
Inventor of the branch
• Alan Turing was an English mathematician, computer
scientist, logician, cryptanalyst, philosopher and
theoretical biologist.
• Turing is widely considered to be the father of
theoretical computer science and artificial
intelligence.
• He proposed an experiment that became known as
the Turing test, an attempt to define a standard for a
machine to be called "intelligent".
– A computer could be said to "think" if a human
interrogator could not tell it apart, through conversation,
from a human being.
photo of Alan turing in 1928 at age 16 (left) and a photo of Benedict
Cumberbatch who portrayed Turing in the 2014 Hollywood
movie The Imitation Game (right)
The birth of Artificial Intelligence (1943 – 1956):
• Warren McCulloch and Walter Pitts in (1943) proposed
a model of an artificial neural network and
demonstrated that simple network structures can learn.
• McCulloch is the second “founding father” of AI after
Alan Turing, created the corner stones of neural
computing and artificial neural networks (ANN).
• The third founder of AI was John von Neumann, a
Hungarian-born mathematician. He encouraged two of
his students to build the first neural network computer
in 1951.
• Claude Shannon: shared Alan Turing’s ideas on the
possibility of machine intelligence.
• In 1950, Shanon published a paper on chess-playing
machines, which pointed out that a typical chess
game involved about 10120 possible moves.
– Even if the new von Neumann-type computer could
examine one move per microsecond, it would take 3* 10106
years to make its first move.
• In 1956, Shanon and two of his colleges organized a
summer workshop for researchers interested in the
study of machine intelligence, artificial neural nets
and automata theory.
• this workshop gave birth to a new science called
Artificial Intelligence (AI).
 The rise of AI (1956 - late 1960s)
• Frank Rosenblatt proved the perceptron convergence
theorem, demonstrating that his learning algorithm
could adjust the connection strengths of a perceptron.
• One of the most ambitious projects of the era of great
expectations was the General Problem Solver (GPS).
– general-purpose program to simulate human-solving methods.
• However, GPS failed to solve complex problems.
– The program was based on formal logic and could generate an
infinite number of possible operators.
– The amount of computer time and memory that GPS required
to solve real-world problems led to the project being
abandoned.
Unfulfilled promises, or the impact of reality
(late 1960s - early 1970s)
• 1970 had witnessed the downfall of AI as Most of
government funding for AI projects was cancelled.
• AI was still a relatively new field, academic in nature,
with few practical applications apart from playing
games.
• To the outsider, the achieved results would be seen
as toys, as no AI system at that time could manage
real-world problems.
• Many of the problems that AI attempted to solve
were too broad and too difficult.
The technology of expert systems
(early 1970s – mid-1980s)
• When researchers finally realized that the only way
to deliver practical results was to solve typical cases
in narrow areas of expertise.
• the DENDRAL project which is supported by NASA
was developed at Stanford University to determine
the molecular structure of Martian soil.
• The project focuses on incorporating the expertise of
an expert into a computer program to make it
perform at a human expert level. Such programs
were later called expert systems.
• DENDRAL marked a major “paradigm shift” in AI: a
shift from general-purpose, knowledge-sparse weak
methods to domain-specific, knowledge intensive
techniques.
• In 1986, in Waterman’s survey reported a remarkable
number of successful expert system applications in
different areas.
– chemistry, electronics, engineering, geology, management,
medicine, process control and military science.
• However, expert system of that time had suffered
from many issues regarding narrow domain of
expertise, difficulty in recognizing domain boundaries,
lack even basic understanding of the domain area and
the inability to learn from their experience.
How to make machine learn, or the rebirth of
neural networks (1980 – onwards)
• In mid-80s, researchers decided to have a new look at
neural networks.
• By the late 60s, most of the basic ideas and concept
necessary for neural computing had already been
formulated. The solution only start to emerge in mid-
80s.
– The major reason for the delay was technological: there
were no PCs or powerful workstations to model and
experiment with ANN.
• The major findings happened in the eighties can be
summarized in the following discoveries:
1. In 1980, Grossberg provided the basis for a new class
of neural networks established using a new self-
organization theory (adaptive resonance theory).
2. In 1982, Hopfield introduced neural networks with
feedback – Hopfield networks, which attracted much
attention in the eighties.
3. In 1982, Kohonen introduced the self-organizing
mapping.
4. In 1983, Barto, Sutton and Anderson introduced
reinforcement learning and its application in control.
5. The real breakthrough came in 1986 when the back-
propagation learning algorithm, first introduced by
Bryson and Ho in 1969, was reinvented by Rumelhart
and McClelland.
ML vs AI
• Artificial Intelligence can be defined as an area of
computer science that has an emphasis on the
creation of intelligent machines that can work and
react like humans.
• Machine Learning can be defined as a subset of AI or
can be termed as an application of Artificial
Intelligence. In Machine Learning, machines have the
ability to learn on their own without being explicitly
programmed.
• It allows applications to modify themselves based on
data in real-time scenarios.
AI vs ML vs DL
Characteristics of an Intelligent System

• A truly intelligent system is characterized by the

following features:
1. exhibit adaptive goal-oriented behavior.
2. learn from experience
3. use vast amounts of knowledge
4. exhibit self awareness
5. interact with humans using language
6. tolerate error and ambiguity in communication
7. respond in real time
IS Applications
• Some applications of intelligent systems can
be found in agriculture, finance, gambling,
medicine, music, robotics, and weather
forecasting.
• Fuzzy logic and Neuro-fuzzy can also be found
in some appliances like video camera, washing
machine, rice cooker, and air-conditioner.
Forms of Intelligent Systems
• ANN can learn, adapt to changes in a problem’s
environment, and establish patterns in situations
where rules are not known.
• However, they lack explanation facilities and usually
act as black box. The process of training neural
networks with current technologies is slow, and
frequent retraining can cause serious difficulties.
• Due to this, another branch of research that attracts
attention is fuzzy logic. As this technology can deal
with vague, imprecise and uncertain knowledge and
data.
• Fuzzy Logic or fuzzy set theory was introduced by
Professor Lotfi Zadeh in 1965.
• Eventually, fuzzy theory, ignored in the West, was
taken seriously in the East – by the Japanese.
• It has been used successfully since 1987 in Japanese-
designed dishwashers, washing machines, air
conditioners, television sets, copiers, and even cars.
• Another form of IS is the Evolutionary or Genetic
algorithms (GA): GAs are a class of stochastic search
algorithms based on biological evolution. Given a
clearly defined problem to be solved and a binary
string representation for candidate solutions.