Pokhara University

Bachelor of Computer Engineering

Lecture Notes
In

Artificial Intelligence
1.Introduction to AI
Intelligence
Intelligence is:
– The ability to reason
– The ability to understand
– The ability to create
– The ability to Learn from experience
– The ability to plan and execute complex tasks
The intelligent behavior may include
– Everyday tasks: recognize a friend, recognize who is calling, translate from one language to
another, interpret a photograph, talk, and cook a dinner
– Formal tasks: prove a logic theorem, geometry, calculus, play chess, checkers, or Go
– Expert tasks: engineering design, medical designers, financial analysis
Artificial Intelligence
AI is the branch of computer science concerned with making computers behave like humans. In other
words, AI is the science and engineering of making intelligent machines, especially intelligent computer
programs. The process may include
- Learning (Gaining of information and rules for using the information)
- Reasoning (Using the rules to reach approximate or definite conclusions)
- Self-Correction
According to Barr and Feigenbaum:
“Artificial Intelligence is the part of computer science concerned with designing intelligence
computer systems, that is, systems that exhibit the characteristics we associate with intelligence in
human behavior.”
According to Elaine Rich:
“AI is the study of how to make computers do things at which, at the moment, people are better”
An AI system should have
- Capability to provide reason about something
- Capability of natural language processing
- Capability of learning past experience
- Capability of self-correction
Views of AI fall into four categories
Thinking humanly Thinking rationally

Acting humanly Acting rationally

- Acting humanly: The Turing Test approach

The Turing Test is a method for determining whether or not a computer is capable of thinking like a
human. The test is named after Alan Turing, an English mathematician who pioneered artificial
intelligence during the 1940s and 1950s, and who is credited with devising the original version of the
test. According to this kind of test, a computer is deemed to have artificial intelligence if it can mimic
human responses under specific conditions.
Consider the following setting. There are two
rooms, A and B. One of the rooms contains a
computer. The other contains a human. The
interrogator is outside and does not know
which one is a computer. He can ask questions
through a teletype and receives answers from
both A and B. The interrogator needs to
identify whether A or B are humans. To pass
the Turing test, the machine has to fool the
interrogator into believing that it is human.
To pass a Turing test, a computer must have following capabilities:
 Natural Language Processing: Must be able to communicate in English
successfully 
 Knowledge representation: To store what it knows and hears. 
 Automated reasoning: Answer the Questions based on the stored
information. 
 Machine learning: Must be able to adapt in new circumstances.
- Thinking humanly: The cognitive modeling approach
 Make the machines with mind.
 Cognition: The action or process of acquiring knowledge and understanding
through thought, experience and senses.
 How do humans think?
Requires scientific theories of internal brain activities (cognitive model).
Once we have precise theory of mind, it is possible to express the theory
as a computer program.
 Two ways of doing this is:
Predicting and testing human behavior (cognitive science)
Identification from neurological data (Cognitive neuroscience)
- Thinking rationally: The “laws of thought approach”
 Aristotle was one of the first who attempt to codify the right thinking that is
irrefutable reasoning process. He gave Syllogisms that always yielded correct
conclusion when correct premises are given.
 For example:
Ram is a man
Man is mortal
i.e. Ram is mortal
These laws of thought were supposed to govern the operation of the
mind; their study initiated a field called logic. The logistic tradition in AI
hopes to create intelligent systems using logic programming.
- Acting rationally : The rational agent approach
 An agent is something that acts.
 Computer agent is expected to have following attributes:
- Autonomous control
- Perceiving their environment
- Persisting over a prolonged period of time
- Adapting to change
- And capable of taking on another’s goal
 Rational behavior: doing the right thing
 The right thing: that which is expected to maximize goal achievement, given the
available information
 Rational Agent is one that acts so as to achieve the best outcome or, when
there is uncertainty, the best expected outcome.
In this approach the emphasis is given to correct inferences.
AI and related fields
Different fields have contributed to AI in the form of ideas, viewpoints and techniques.
Philosophy:
Logic, reasoning, mind as a physical system, foundations of learning, language and
rationality.
Mathematics:
Formal representation and proof algorithms, computation, undesirability, intractability,
probability.
Psychology:
Adaptation, phenomena of perception and motor control.
Economics:
Formal theory of rational decisions, game theory.
Linguistics:
Knowledge representation, grammar
Neuroscience:
Physical substrate for mental activities
Control theory:
Homeostatic systems, stability, optimal agent design
Brief History of AI
The term “Artificial Intelligence” was used for the first time in 1956 by an American scientist John McCarthy
who is referred to as the Father of AI. McCarthy also come up with a programming language called LISP (i.e.
List-Processing), which is still used to program computer in AI that allow the computer to learn.
Further, the major achievements can be listed as below:
1943 First electronic computer “Colossus” was developed.
1949 First commercial stored program computer was developed.
- Alan Turing proposes the Turing test as a measure of machine intelligence.
1950 - Claude Shannon published a detail analysis of chess playing as search.
- Isaac Asimov published his three laws of Robotics.
The first working AI programs were written to run on the Ferranti Mark machine of the University
1951 of Manchester; a checkers-playing program written by Christopher Stavechey and a chess-playing
program is written by Dietrich Prinz
The first Dartmouth college summer AI conference is organized by John McCarthy, Marvin
1955
Minsky, Nathan Rochester of IBM and Claude Shannon.
- The name artificial intelligence is used for the 1st time as the topic of the second Dartmouth
Conference, organized by John McCarthy.
1956
- The first demonstration of the Logic Theorist (LT) written by Allen Newell, J.C. Shaw and
Merbart Simon pus is called the first AI program
1957 The general problem Solver (GPS) demonstrated by Newell, Shaw and Simon
1958 John McCarthy at MIT invented the Lisp Programming Language.
- John McCarthy and Marvin Minsky founded the MIT AI Lab.
1959
- First industrial robot company, animation was established.
1972 Prolog programming language was developed by Alain Colmerauer
First National Conference of the American Association for Artificial Intelligence (AAAI) was held
1980
at Stratford.
Mid 1980’s Neural networks become widely used with the Back propagation algorithm.
AI system exist in real environments with real sensory inputs (i.e. Intelligent
1994
Agents)
1997 First time AI system controlled a spacecraft named “Deep Space II”
2007 Checkers is solved by a team of researchers of the University of Alberta.
Programmers are still trying to develop a computer which can successfully pass the
Present
“Turing Test”.

Importance and Application of AI

Artificial intelligence has been used in a wide range of fields including medical diagnosis, stock
trading, robot control, law, remote sensing, scientific discovery and toys. Many thousands of AI
applications are deeply embedded in the infrastructure of every industry. In the late 90s and early 21st
century, AI technology became widely used as elements of larger systems, but the field is rarely credited
for these successes.
Game Playing
Machines can play master level chess. There is some AI in them, but they well against people
mainly through brute force copulation, looking at hundreds of thousands of positions.
Speech Recognition
It is possible to instruct some computers using speech. In 1990s, computer speech recognition
reached a practical level for limited purposes.
Understanding Natural Language
To perform many natural language processing tasks such as machine translation, summarization,
information extraction, word sense disambiguation need the AI in machine.
Computer Vision
Computer vision is concerned with the theory behind artificial system that extract information
from images. The image data can take many forms such as videos sequences views from multiple
cameras and data from a medical scanner. Application range from simple tasks such as industrial
machine, vision system which count bottles speeding by on a production line to research into
artificial intelligence and computers or robots that can comprehended the world around them.
Expert System
Expert system need the AI to perform its task. One of the first expert system was MYCIN in 1974
which diagnosis bacterial infections of the blood and suggests treatments. It did better that makes
medical students practicing doctors provided to limitations were observed.
Finance
Financial institutions have long used artificial neural network systems to detect charges or claims
outside of the norm, flagging these for human investigation. Use of AI in banking can be tracked
back to 1987 when Security Pacific National Bank in USA set-up a Fraud Prevention Task force
to counter the unauthorized use of debit cards.
Hospitals and medicine
Artificial neural networks are used as clinical decision support systems for medical diagnosis,
such as in Concept Processing technology in EMR software.
Other tasks in medicine that can potentially be performed by artificial intelligence include:
 Computer-aided interpretation of medical images. Such systems help scan digital
images, e.g. from computed tomography, for typical appearances and to highlight
conspicuous sections, such as possible diseases. A typical application is the detection of
a tumor.
 Heart sound analysis
 Companion robots for the care of the elderly
Heavy industry
Robots have become common in many industries. They are often given jobs that are considered
dangerous to humans. Robots have proven effective in jobs that are very repetitive which may
lead to mistakes or accidents due to a lapse in concentration and other jobs which humans may
find degrading. Japan is the leader in using and producing robots in the world. In 1999, 1,700,000
robots were in use worldwide.

Online and telephone customer service

Artificial intelligence is implemented in automated online assistants that can be seen
as avatars on web pages. It can avail for enterprises to reduce their operation and
training cost. A major underlying technology to such systems is natural language processing.
Toys and games
The 1990s saw some of the first attempts to mass-produce domestically aimed types of basic
Artificial Intelligence for education, or leisure. This prospered greatly with the Digital
Revolution, and helped introduce people, especially children, to a life of dealing with various
types of Artificial Intelligence. AI has also been applied to video games, for example video game
bots, which are designed to stand in as opponents where humans aren't available or desired
Music
The evolution of music has always been affected by technology. With AI, scientists are trying to
make the computer emulate the activities of the skillful musician. Composition, performance,
music theory, sound processing are some of the major areas on which research in Music and
Artificial Intelligence are focusing.
Aviation
The Air Operations Division (AOD) uses AI for the rule based expert systems. The AOD has use
for artificial intelligence for replacement operators for fighting and training simulators, mission
management aids, support systems for tactical decision making, and post processing of the
simulator data into symbolic summaries.
Knowledge and Learning
Knowledge is the information about a domain that can be used to solve problems in that domain. To
solve many problems requires much knowledge, and this knowledge must be represented in the
computer. As part of designing a program to solve problems, we must define how the knowledge will be
represented.
A representation scheme is the form of the knowledge that is used in an agent. A representation of some
piece of knowledge is the internal representation of the knowledge. A representation scheme specifies the
form of the knowledge. A knowledge base is the representation of all of the knowledge that is stored by
an agent.
A good representation scheme is a compromise among many competing objectives. A representation
should be
 Rich enough to express the knowledge needed to solve the problem.
 Willing for efficient computation
 Able to be acquired from people, data and past experiences.
Knowledge is the body of facts and principles. Knowledge can be language, concepts, procedures, rules,
ideas, abstractions, places, customs, and so on. (Study of knowledge is called Epistemology)
Types of knowledge
The types of knowledge include procedural knowledge, declarative knowledge and heuristic
knowledge.
- Meta Knowledge
It is a knowledge about a knowledge and how to gain them.
- Procedural knowledge
Procedural knowledge is compiled or processed form of information. Procedural knowledge is
related to the performance of some task. For example, sequence of steps to solve a problem is
procedural knowledge.
- Declarative knowledge
Declarative knowledge is passive knowledge in the form of statements of facts about the world.
For example, mark statement of a student is declarative knowledge.
- Heuristic knowledge
Heuristics knowledge are rules of thumb or tricks. Heuristic knowledge is used to make
judgments and also to simplify solution of problems. It is acquired through experience. An expert
uses his knowledge that he has gathered due to his experience and learning.
- Structural Knowledge
Describes what relationship exists between concepts/ objects.
Learning:
Learning is acquiring new or modifying existing knowledge, behaviors, skills, values and may involve
synthesizing different types of information.
Machine learning, a branch of AI, is a scientific discipline concerned with the design and development of
algorithms that allow computers to evolve behaviors based on empirical data such as from sensor data or
database.
Inferential Knowledge
Natural language understanding requires inference i.e., assumptions about what is typically true of the
objects or situations under consideration. Such information can be coded in structures known as frames.
- Need of Frames
Frame is a type of diagram used in many AI applications including vision and natural
language processing. Frames provide a convenient structure for
representing objects that are typical to a conventional situations. The
situations to represent may be visual scenes, structure of complex physical objects, etc.
A frame is similar to a record structure and corresponding to the fields and values are
slots and slot fillers. Frame systems usually have collection of frames connected to each
other. Value of an attribute of one frame may be another frame.
A frame for a book is given below.

Slots Fillers
publisher Thomson
title Expert Systems
author Giarratano
edition Third
year 1998
pages 600

Intelligent agents and its type and performance measures

An Intelligent Agent perceives it environment via sensors and acts rationally upon that environment
with its effectors (actuators). Hence, an agent gets percepts one at a time, and maps this percept
sequence to actions.
Properties
– Autonomous
– Interacts with other agents plus the environment
– Reactive to the environment
– Pro-active (goal- directed)
Types of Agent
- Simple Reflex Agent
 They choose actions only
based on the current
percept.
 They are rational only if a
correct decision is made
only on the basis of current
precept.

 Their environment is completely observable

- Model Based Agents (Reflex Agent with
Internal State)
 They use a model of the world to
choose their actions. They maintain an
internal state.
 Model − the knowledge about how the
things happen in the world.
 Internal State − It is a representation of unobserved aspects of current state depending
on percept history.
- Goal Based Agents
They choose their actions in order to
achieve goals. Goal-based approach is
more flexible than reflex agent since the
knowledge supporting a decision is
explicitly modeled, thereby allowing for
modifications.
- Utility Based Agents
 They choose actions based on a
preference utility for each state.
Goals are inadequate when −There
are conflicting goals, out of which
only few can be achieved.
 Goals have some uncertainty of
being achieved and you need to
weigh likelihood of success against the importance of a goal.
PEAS
To design a rational agent we must specify its task environment. Standing for performance, environment,
actuators and sensors, PEAS define task environments about formulating the performance of intelligent
agents.

Q. Point out the task of designing an automated taxi driver according to PEAS description.
 Performance measure: Safe, fast, legal, comfortable trip, maximize profits
 Environment: Roads, other traffic, pedestrians, customers
 Actuators: Steering wheel, accelerator, brake, signal, horn
 Sensors: Cameras, sonar, speedometer, GPS, odometer, engine sensors and keyboard

Q. Point out the task of designing a Medical diagnosis system according to PEAS description
• Performance measure: Healthy patient, minimize costs, lawsuits
• Environment: Patient, hospital, staff
• Actuators: Screen display (questions, tests, diagnoses, treatments, referrals)
• Sensors: Keyboard (entry of symptoms, findings, patient's answers)

Q. Machines can be made intelligent artificially but ultimately persons make the machines. So
who is more intelligent - the artificial machine or the person? Discuss
Ans: Human have done considerable work in designing a machine but the machine may not need to do
very much to operate well. An example is, thermostat. It is difficult to design a thermo stat so that it
turns on and off at exactly the right temperature but the thermostat itself does not have to do more
computations.
All the logic behind making the machine specifies what needs to be mechanized and how to be
machinated but not in vice versa. The AI reasoning in human involves all the possibilities to
determine how to make a complete machine. The natural intelligence of human made the AI which
may not cope on real time. Hence, the above point are supportive on the favor of the humans.
What can AI systems do and don’t do
Today’s successful AI systems operate in well-defined domains and employ narrow, specialized
knowledge. Common sense knowledge is needed to function in complex, open-ended worlds. Such a
system also needs to understand unconstrained natural language. However these capabilities are not yet
fully present in today’s intelligent systems.
What can AI systems do What can AI systems NOT do yet?
Today’s AI systems have been able to achieve • Understand natural language robustly (e.g., read
limited success in some of these tasks. and understand articles in a newspaper)
• In Computer vision, the systems are capable of • Surf the web
face recognition • Interpret an arbitrary visual scene
• In Robotics, we have been able to make vehicles • Learn a natural language
that are mostly autonomous. • Construct plans in dynamic real-time domains
• In Natural language processing, we have • Exhibit true autonomy and intelligence
systems that are capable of simple machine
translation.
• Today’s Expert systems can carry out medical
diagnosis in a narrow domain
• Speech understanding systems are capable of
recognizing several thousand words continuous
speech