Transcript AIIntro

CPTR 314
Introduction to
Artificial Intelligence
Instructor: Dr. Eduardo
Urbina
Artificial Intelligence Definition


AI may be defined as the branch of computer
science that is concerned with the automation of
intelligent behavior
Artificial Intelligence uses the techniques of
Computer Science
–
–
–
Data Structures for knowledge representation
Programming Languages & Techniques
Algorithms
What is Intelligence?





Does it mean to perform operations fast?
Is it learned?
What is the relationship with intuition?
What is the relationship with creativity?
Can machines be intelligent? If so, Is it
morally correct? Can only God create
intelligence?
Definition

AI is the collection of problems and
methodologies studied by artificial
intelligence researchers
How can you model Intelligence?

Mathematical logic
–

Main technique so far
Connectionist Networks
–
De-emphasizes logic and the functioning of the
rational mind but concentrates in the architecture
of the physical mind
How do we model Intelligence?

Using artificial life and genetic algorithms
–
–
Applies the principles of biological evolution to the
problems of finding solutions to difficult problems
Solutions are are found by analyzing competing
solutions; solutions with promise will tend to
survive
How do we model Intelligence?


Social systems provide another metaphor for
intelligence; the basis of this are agents
An agent is an element of a society that can
perceive aspects of its environment and
affect that environment either directly or
through cooperation with other agents
Agents Characteristics

Agents are autonomous or semi-autonomous
–



Each agent has little or no knowledge of what other agents
do or how they do things
Agents are sensitive to their own surroundings
Agents are interactional; they cooperate on a
particular task
Intelligence is the result of the society as a whole not
just a property of an individual agent
Overview of AI

Most fundamental concerns of AI
researchers are
–
–
Knowledge representation and search
Machine learning
AI is Complex

Mimicing human intelligence is hard
–
–
–
Use of symbolic reasoning
Humans use inexact, missing or poorly defined
information; they rely on intuition.
Answers may not be exact or optimal but just
sufficient to solve a problem
AI Sub-disciplines






Game Playing
Automated Reasoning
and Theorem Proving
Expert Systems
Natural Language
Robotics
Languages and
Environments



Machine Learning
Parallel Distributed
Processing
Modeling Human
Performance
AI features




Use of computers to do symbolic reasoning, pattern
recognition, learning, or some other form of
inference
A focus of problems that do not respond to
algorithmic solutions
A concern with problem solving using inexact,
missing, or poorly defined information
Provide answers that are neither exact nor optimal,
but in some sense “sufficient”