What is Artificial Intelligence? AI, Artificial Intelligence Many definitions No universal agreement

advertisement
What is Artificial Intelligence?
AI, Artificial Intelligence
Many definitions
No universal agreement
 Based on different point of view, a different
definition

AI & Expert Systems
AI & Expert Systems
Are computers the right kind
of machine to be made
intelligent?
Computers can be programmed to
simulate any kind of machine.
AI & Expert Systems
Are computers fast enough to
be intelligent?
My own opinion is that the computers of 30
years ago were fast enough if only we
knew how to program them. Of course,
quite apart from the ambitions of AI
researchers, computers will keep getting
faster.
AI & Expert Systems
Does AI aim to put the human
mind into the computer?
Some researchers say they have that
objective, but maybe they are using the
phrase metaphorically. The human mind
has a lot of peculiarities, and I'm not sure
anyone is serious about imitating all of
them.
AI & Expert Systems
Different definitions of AI
Thought
Behavior
Human
AI & Expert Systems
Ideal
???
What is artificial intelligence (AI)?
Why study AI?
How can you tell whether a machine
or program is intelligent?
What kinds of things (if any) can be
learned by a machine?
When can (should) machines replace
human experts?
AI & Expert Systems
What is AI?
AI definition was always debatable since the
meaning of intelligence was not well-defined.
Intelligence might be defined as the capacity to
acquire and apply knowledge.
So AI was defined as:


AI is the study of ideas that enable computers to be
intelligent.
AI is the part of computer science concerned with
design of computer systems that exhibit human
intelligence.
AI & Expert Systems
What is AI?
From the above two definitions, we can see
that AI has two major roles:
Study the intelligent part concerned with
humans.
 Represent those actions using computers.

AI & Expert Systems
Goals of AI
To make computers more useful by letting
them take over dangerous or tedious tasks
from human
Understand principles of human intelligence
AI & Expert Systems
More explanation
Thought VS behavior
Ideal?
Rational, reasonable and correct
 Humans sometimes do incorrect things


Humans are sometimes irrational
AI & Expert Systems
Human-like VS Rational?
A system thinks or acts like a human
Methods employed by humans
 But, sometimes we even don’t know
how??!!

A system thinks or acts rationally
Don’t care the methods, only care about
the correctness of the result
 Usually employ mathematical and logical
models

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Acting humanly
The Turing Test approach

a human questioner cannot tell if


there is a computer or a human answering his
question, via teletype (remote communication)
The computer must behave intelligently
Intelligent behavior

to achieve human-level performance in all
cognitive tasks
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Acting humanly
These cognitive tasks include:

Natural language processing


Knowledge representation


to store information effectively & efficiently
Automated reasoning


for communication with human
to retrieve & answer questions using the stored
information
Machine learning

to adapt (適應) to new circumstances
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The total Turing Test
Includes two more issues:

Computer vision


to perceive objects (seeing)
Robotics

to move objects (acting)
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Thinking humanly
The cognitive modeling approach
To enable computer thinking like a human
 We must have a precise theory of mind
 Cognitive Science is dealing with the
theory of mind


bringing together the theories of AI and
psychology
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Thinking rationally
The laws of thought approach
Logic is the laws of thought to govern the
operation of mind.
 Usually logic takes as inputs several
premises and deduces a conclusion.
 Example: If X and Y then Z (X ^ Y => Z)

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Acting rationally
The rational agent approach
Given one’s beliefs (facts), one acts in
order to achieve one’s goals
 An agent is just something that perceives
(感應) and acts


like a program: get inputs, and perform some
outputs
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Acting rationally
Logic  only part of a rational agent, not all
of rationality
Sometimes logic cannot reason a correct
conclusion
 At that time, some specific (in domain) human
knowledge or information is used

Thus, it covers more generally different
situations of problems

Compensate the incorrectly reasoned
conclusion
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Acting rationally
Study AI as rational agent – 2 advantages:

It is more general than using logic only


Because: LOGIC + Domain knowledge
It allows extension of the approach with more
scientific methodologies
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The Foundation of AI
Philosophy
At that time, the study of human intelligence
began with no formal expression
 Initiate the idea of mind as a machine and its
internal operations

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The Foundation of AI
Mathematics formalizes the three main
area of AI: computation, logic, and
probability

Computation leads to analysis of the
problems that can be computed

complexity theory
Probability contributes the “degree of belief”
to handle uncertainty in AI
 Decision theory combines probability theory
and utility theory (bias)

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The Foundation of AI
Human
thinking
Psychology
How do humans think and act?
 The study of human reasoning and acting
 Provides reasoning models for AI
 Strengthen the ideas


humans and other animals can be considered
as information processing machines
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The Foundation of AI
Computer Engineering
How to build an efficient computer?
 Provides the artifact (工具) that makes AI
application possible
 The power of computer makes computation
of large and difficult problems more easily
 AI has also contributed its own work to
computer science, including: time-sharing,
the linked list data type, OOP, etc.

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The Foundation of AI
Control theory and Cybernetics
How can artifacts operate under their own
control?
 The artifacts adjust their actions

To do better for the environment over time
 Based on an objective function and feedback
from the environment


Not limited only to linear systems but also
other problems

as language, vision, and planning, etc.
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The Foundation of AI
Linguistics

For understanding natural languages

different approaches has been adopted from
the linguistic work
Formal languages
 Syntactic and semantic analysis
 Knowledge representation

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The history of AI
Read it yourselves, if you feel interested
in it
In short
AI has been born for over 50 years
 Many different areas arise during this time
 AI has evolved from laboratory research to
an industry

AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
The state of the art
What can AI do today?
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Autonomous Planning & Scheduling:

Autonomous rovers.
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Autonomous Planning & Scheduling:

Telescope scheduling:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Autonomous Planning & Scheduling:

Analysis of data:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Medicine:

Image guided surgery
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Medicine:

Image analysis and enhancement
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Transportation:

Autonomous vehicle control:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Transportation:

Pedestrian detection:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Games:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Games:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Robotic toys:
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
AI Applications
Other application areas:
Bioinformatics:


Gene expression data analysis
Prediction of protein structure
Text classification, document sorting:


Web pages, e-mails
Articles in the news
Video, image classification
Music composition, picture drawing
Natural Language Processing .
Perception.
AI & Expert Systems
© Dr. Khalid Kaabneh, AAU
Download