Transcript AI Introduction

Artificial Intelligence
CS105
Group Presentation Project
• Week of November 7 to November 11
• Group presentation
• Groups of 3/4
–
–
–
–
Each group will get 12/16 minutes maximum
Each member must speak for 4 minutes
Q&A session for 3/4 minutes
Team Leaders
• Topics:
1. An existing AI application(s)
2. An AI application that will possibly be created in the future and
you would invest money in it
3. Technology in an AI movie – their advantages & disadvantages
• Questions: What is it? How does it work?
– Don’t have to explain any technical details if you don’t want to
Group Presentation Project
• Evaluation:
–
–
–
–
Content and details
Organization
Member evaluation form
Dress up on the day of presentation
• Powerpoint files should be received BEFORE the beginning of
class.
• Attendance is mandatory this week
Finding your group!
Artificial Intelligence
Artificial Intelligence
• Study and design of machines that can think like a
human being
– Attempt to understand human intelligence
• But how do you define intelligence?
– Abstract thought, understanding, communication,
reasoning and learning, planning and problem solving
• Goal: understand primary issues involved and the
challenges
Artificial Intelligence
• Intelligent machines
List
250099221
675490823
945784675
232343987
121108908
232308738
686532900
124345072
212576899
.
.
Turing Test
• A behavioral test to determine whether a
computer system is intelligent
• Natural language processing
• Knowledge about various topics
• Reasoning
Weak equivalence: The equality of two
systems based on their results
Strong equivalence: The equality of two
systems based on their results and the
process by which they arrive at those
results
State of the art
Deep Blue beat Gary Kasparov,
world champion in 1997 in a set
of six games by two wins
Search engines
Tartan Racing: 2007 DARPA
Urban challenge winner.
Autonomous vehicle capable of
driving in traffic.
Mobile humanoid
robots: MIT media labs
Responsive
environments:
Bill Gates home
TD Gammon learned how
to play backgammon
Foundations of AI
• Psychology
– How do humans and animals think and act?
• Computer Science/Engineering
– How to build an efficient computer system?
• Linguistics
– How does language relate to thought?
•
•
•
•
•
Philosophy
Mathematics
Economics
Neuroscience
Education / Learning
Knowledge Representation
• Many ways
– Natural language : not efficient for processing
– Mathematical notation: rigorous
– Want: Logical view so we can process it efficiently
• Capture facts and relationships
Semantic network
• A graph or network that presents meaningful
relationships among objects/concepts
has
Cat
Vertebra
Fur
is a
has
is a
Animal
has
is a
Mammal
Bear
is a
is a
lives in
Fish
lives in
Water
Whale
Search Trees
• Hierarchical structures that represents the
alternatives of situations/choices
– Nodes: Points in a tree that contains some form of
data
– Edges: Link between nodes in a search tree
– Game tree: is a graph whose nodes are the
positions in a game and whose edges are moves
Search Trees
P
Edges
Nodes
7
9
2
Parent nodes
Child nodes
Leaf nodes
P
C
P
C
5
4
P
C
8
3
P
C
6
P
C
L
0
P
C
L
12
4
7
L
L
L
3
5
L
L
6
1
L
L
Game Trees
Depth first
Breadth
first search
search
Player 1
Player 2
Wins!
Wins!
Player 1
Player 2
Wins!
Wins!
Thought Question
Think of five questions that you might issue as
the interrogator of a Turing test. Why would a
computer have difficulty answering them
well? Please explain your reasoning.