Unit - 1 : INTRODUCTION
What is Intelligence?
Something that characterizes humans and show rational behavior or best behavior is
intelligence. It is ability to apply knowledge in order to perform better in an environment.
• 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
Human mind:
• Can use common sense and past experience for problem solving.
• Can understand natural language processing.
• Can percept environment like see, hear
• Can learn from experiences and acquire knowledge
• Can reason about facts and deduce new facts.
What is Artificial Intelligence?
Artificial Intelligence is a branch of Science which deals with helping machines to find
solutions to complex problems in a more human-like fashion. This generally involves
borrowing characteristics from human intelligence, and applying them as algorithms in a
computer friendly way. A more or less flexible or efficient approach can be taken depending on
the requirements established, which influences how artificial the intelligent behavior appears.
• “Giving machines ability to perform tasks normally associated with human intelligence.”
• Branch of computer science that aims to create intelligence of machines • Part of
computer science concerned with designing intelligent machines
• Deals with issues like inference, reasoning, problem solving, knowledge representation,
planning, natural language process ing, perceptron, etc.
AI is generally associated with Computer Science, but it has many important links with other
fields such as Math, Psychology, Cognition, Biology and Philosophy, among many others.
Our ability to combine knowledge from all these fields will ultimately benefit our progress in
the quest of creating an intelligent artificial being.
AI was coined by John McCathy in 1956 who defines AI as science and engineering of
making intelligent machines especially intelligent computers.
Different definitions of AI are given by different books/writers. These definitions can be
divided into two dimensions.
System that thinks like humans System that think rationally
System that acts like humans System that act rationally
Top dimension is concerned with thought processes and reasoning, where as bottom
dimension addresses the behavior.
The definition on the left measures the success in terms of fidelity of human
performance, whereas definitions on the right measure an ideal concept of intelligence,
which is called rationality.
Human-centered approaches must be an empirical science, involving hypothesis and
experimental confirmation. A rationalist approach involves a combination of mathematics
and engineering.
Acting Humanly: The Turing Test Approach
• proposed by Alan Turing (1950)
• designed to convince the people that whether a particular machine can think or not • a
test based on indistinguishability from undeniably intelligent entities- human beings
• Involves an interrogator who interacts with one human and one machine. Within the given
time the interrogator has to find out which one is human and which one is machine
• The computer passes the test if a human interrogator can not tell whether the written
response come from human or not
• To pass a Turing test, a computer must have following capabilities:
➢ Natural language processing
➢ Knowledge representation
➢ Automated reasoning
➢ Machine learning
Turing test avoid the physical interaction with human interrogator. Physical simulation
of human beings is not necessary for testing the intelligence.
The total Turing test includes video signals and manipulation capability so that the
interrogator can test the subject’s perceptual abilities and object manipulation
ability. To pass the total Turing test computer must have following additional
capabilities:
➢ Computer Vision: To perceive objects
➢ Robotics: To manipulate objects and move
Chinese room argument
• Searle (1999) summarized the Chinese Room argument
• Imagine a native English speaker who knows no Chinese locked in a room full of boxes
of Chinese symbols (a data base) together with a book of instructions for manipulating
the symbols (the program).
• Imagine that people outside the room send in other Chinese symbols which, unknown to
the person in the room, are questions in Chinese (the input).
• And imagine that by following the instructions in the program the man in the room is able
to pass out Chinese symbols which are correct answers to the questions (the output).
• The program enables the person in the room to pass the Turing Test for understanding
Chinese but he does not understand a word of Chinese.
Thinking Humanly: Cognitive approach
Make the machines with mind. Cognition means the action or process of acquiring knowledge
and understanding through thought, experience and senses. To make a machine that think like
human brain, scientific theories of internal brain activities (cognitive model) are required. Two
ways of doing this is:
– Predicting and testing human behavior (cognitive science)
– Identification from neurological data (Cognitive neuroscience)
Once we have precise theory of mind, it is possible to express the theory as a computer
program. But unfortunately until up to now there is no precise theory about thinking process of
human brain. Therefore it is not possible to make the machines that think like human brain
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
->Ram is mortal
Let
p(x) -> x is man
q(x) -> x is mortal
then above statement can be written as
p(x) => q(x) Man is mortal
p(Ram) Ram is man
Then from modus ponens q(x) ia also true. That is Ram is mortal
This study initiated the field of logic. The logicist tradition in AI hopes to create
intelligent systems using logic programming.
Problems:
It is not easy to take informal knowledge and state in the formal terms required by
logical notation, particularly when knowledge is not 100% certain.
Solving problem principally is different from doing it in practice. Even problems with
certain dozens of fact may exhaust the computational resources of any computer unless it
has some guidance as which reasoning step to try first.
Acting Rationally: The rational Agent approach:
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 means doing the right thing. The right thing is 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.
One way to act rationally is to reason logically to the conclusion and act on that conclusion.
On the other hand there are also some ways of acting rationally that can not be said to involve
inference. For Example, recoiling from a host stove is a reflex action that is usually more
successful than a slower action taken after careful deliberation.
Advantages: It is more general than laws of thought approach, because correct inference is
just one of several mechanisms for achieving rationality.
It is more amenable/ controllable to scientific development than are approaches based on
human behavior or human thought because the standard of rationality is clearly defined and
completely general.
Applications of AI
Game playing
You can buy machines that can play master level chess for a few hundred dollars. There is some
AI in them, but they play well against people mainly through brute force computation--looking
at hundreds of thousands of positions. To beat a world champion by brute force and known
reliable heuristics requires being able to look at 200 million positions per second.
Speech recognition
In the 1990s, computer speech recognition reached a practical level for limited purposes. Thus
United Airlines has replaced its keyboard tree for flight information by a system using speech
recognition of flight numbers and city names. It is quite convenient. On the the other hand,
while it is possible to instruct some computers using speech, most users have gone back to the
keyboard and the mouse as still more convenient.
Understanding natural language
Just getting a sequence of words into a computer is not enough. Parsing sentences is not
enough either. The computer has to be provided with an understanding of the domain the text
is about, and this is presently possible only for very limited domains.
Computer vision
The world is composed of three-dimensional objects, but the inputs to the human eye and
computers' TV cameras are two dimensional. Some useful programs can work solely in two
dimensions, but full computer vision requires partial three-dimensional information that is not
just a set of two-dimensional views. At present there are only limited ways of representing three
dimensional information directly, and they are not as good as what humans evidently use.
Expert systems
A ``knowledge engineer'' interviews experts in a certain domain and tries to embody their
knowledge in a computer program for carrying out some task. How well this works depends on
whether the intellectual mechanisms required for the task are within the present state of AI.
When this turned out not to be so, there were many disappointing results. One of the first expert
systems was MYCIN in 1974, which diagnosed bacterial infections of the blood and suggested
treatments. It did better than medical students or practicing doctors, provided its limitations
were observed. Namely, its ontology included bacteria, symptoms, and treatments and did not
include patients, doctors, hospitals, death, recovery, and events occurring in time. Its
interactions depended on a single patient being considered. Since the experts consulted by the
knowledge engineers knew about patients, doctors, death, recovery, etc., it is clear that the
knowledge engineers forced what the experts told them into a predetermined framework. In the
present state
of AI this has to be true. The usefulness of current expert systems depends on their users
having common sense.
Heuristic classification
One of the most feasible kinds of expert system given the present knowledge of AI is to put
some information in one of a fixed set of categories using several sources of information. An
example is advising whether to accept a proposed credit card purchase. Information is available
about the
owner of the credit card, his record of payment and also about the item he is buying and about
the establishment from which he is buying it (e.g., about whether there have been previous
credit card frauds at this establishment).
Fairness and Bias in Artificial Intelligence
Fairness and bias in artificial intelligence (AI) are critical issues that have gained significant
attention in recent years. As AI systems are increasingly being used in various domains and
applications, it is crucial to ensure that these systems are fair, unbiased, and equitable.
The bias in AI can be defined as the error that leads to unfair decisions. This can also be called
machine learning bias or algorithmic bias and this can happen due to various sources such as
data collection, algorithm design, human biases, etc.
Fairness in artificial intelligence can be defined as an attempt to correct the algorithmic bias such as
race or ethnicity etc in an automated decision-making process. It is very important to address these
algorithmic biases and mitigate them to follow the ethics of AI.
This can be tricky and subjective at times because fairness can mean different to different people,
and it is very difficult to jump to conclusions about it because fairness has different definitions.
Fairness is a crucial concept and can’t be ignored.
What is control theory in AI?
In AI, control theory is the study of how agents can best interact with their environment to achieve a
desired goal. The goal of control theory is to design algorithms that enable agents to make optimal
decisions, while taking into account the uncertainty of the environment.
One of the key ideas in control theory is the concept of a Markov decision process (MDP). An MDP
is a mathematical model of a decision-making problem in which an agent must choose between a set
of actions in order to maximize its expected reward. MDPs are often used to model problems in
reinforcement learning, where an agent is learning to optimize its behavior by trial and error.
Another important concept in control theory is the idea of a Bellman equation. The Bellman
equation is a mathematical relationship that describes how the value of a particular state in an MDP
depends on the values of the states that can be reached from it. The Bellman equation is used to
solve MDPs by finding the optimal policy, which is the set of actions that maximizes the expected
reward for the agent.
Foundations of AI
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, undecidability,
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
Sociology: Social behavior and social values, traditions
Brief history of AI
What happened after WWII?
– 1943: Warren Mc Culloch and Walter Pitts: a model of artificial boolean neurons to
perform computations.
– First steps toward connectionist computation and learning (Hebbian learning). – Marvin
Minsky and Dann Edmonds (1951) constructed the first neural network computer – 1950:
Alan Turing’s “Computing Machinery and Intelligence”
– First complete vision of AI.
The birth of AI (1956)
- Dartmouth Workshop bringing together top minds on automata theory, neural nets and
the study of intelligence.
– Allen Newell and Herbert Simon: The logic theorist (first nonnumeric thinking program
used for theorem proving)
– For the next 20 years the field was dominated by these participants.
Great expectations (1952-1969)
– Newell and Simon introduced the General Problem Solver.
– Imitation of human problem-solving
– Arthur Samuel (1952-)investigated game playing (checkers ) with great
success. – John McCarthy(1958-) :
– Inventor of Lisp (second-oldest high-level language)
– Logic oriented, Advice Taker (separation between knowledge and reasoning)
– Marvin Minsky (1958 -)
– Introduction of microworlds that appear to require intelligence to solve: e.g.
blocks-world. – Anti-logic orientation, society of the mind.
Collapse in AI research (1966 - 1973)
– Progress was slower than expected.
– Unrealistic predictions.
– Some systems lacked scalability.
– Combinatorial explosion in search.
– Fundamental limitations on techniques and representations.
– Minsky and Papert (1969) Perceptrons.
AI revival through knowledge-based systems (1969-1970)
– General-purpose vs. domain specific
- E.g. the DENDRAL project (Buchanan et al. 1969)
First successful knowledge intensive system.
– Expert systems
- MYCIN to diagnose blood infections (Feigenbaum et al.)
- Introduction of uncertainty in reasoning.
– Increase in knowledge representation research.
- Logic, frames, semantic nets, …
AI becomes an industry (1980 - present)
– R1 at DEC (McDermott, 1982)
– Fifth generation project in Japan (1981)
– American response …
Puts an end to the AI winter.
Connectionist revival (1986 - present): (Return of Neural Network)
– Parallel distributed processing (RumelHart and McClelland, 1986); backprop.
AI becomes a science (1987 - present)
– In speech recognition: hidden markov models
– In neural networks
– In uncertain reasoning and expert systems: Bayesian network formalism
The emergence of intelligent agents (1995 - present)
– The whole agent problem: “How does an agent act/behave embedded in real environments
with continuous sensory inputs”