Revised 10/02
1
of
11
University of Wisconsin
-
Whitewater
Curriculum Proposal Form #3
New Course
Effective Term:
2101 (Spring 2010)
Subject Area
-
Course Number:
COMPSCI 3
70
Cross
-
listing:
N
/
A
(See Note #1 below)
Course Title:
(Limited to 65 char
acters)
Introduction to
Artificial Intelligence
25
-
Character Abbreviation:
AI
Sponsor(s):
Dr. Hien Nguyen, Assistant Professor of Computer Sciences.
Department(s):
Mathematical and Computer Scienc
es
College(s):
Letters and Sciences
Consultation took place
:
NA
Yes (list departments and attach consultation sheet)
Departments:
Multimedia Digital Arts
Programs Affected:
NA
Is paperwork
complete for those programs?
(Use "Form 2" for Catalog & Academic Report updates)
NA
Yes
will be at future meeting
Prerequisites:
MCS 220
or 202 (in Programming thread of new gaming majo
r)
Grade Basis:
Conventional Letter
S/NC or Pass/Fail
Course will be offered:
Part of Load
Above Load
On Campus
Off Campus
-
Location
College:
Letters and Sciences
Dept/Area(s):
Math
.
and Computer Sciences
Instructor:
Dr. Hien Nguyen
Note: If the course is dual
-
listed, instructor
must
be a member of Grad Faculty.
Check if the Course is to Meet An
y of the Following:
Computer Requirement
Writing Requirement
Diversity
General Education Option:
Select one:
Note: For the Gen Ed option, the proposal should address h
ow this course relates to specific core courses, meets the goals of General Education
in providing breadth, and incorporates scholarship in the appropriate field relating to women and gender.
Credit/Contact Hours:
(per semester)
Total lab hours:
0
Total lecture hours:
4
8
Number of credits:
3
Total contact hours:
4
8
Can course be taken more than once for credit? (Repeatability)
No
Yes If "Yes", ans
wer the following questions
:
No of times in major:
No of credits in major:
No of times in degree:
No of credits in degree:
Revised 10/02
2
of
11
Proposal Information:
(Procedures can be found at
http://acadaff.uww.edu/Handbook/Procedures
-
Form3.htm
)
Course justification:
This c
our
s
e will
strengthen
the Math
M
ajor Computer Emphasis,
and
the
Computer Science minor.
It
is also needed in the
Technology track
of the newly proposed
gaming major
in c
ollaboration with the
College
of Art
s and Communications. This course
is also
very valuable for students from Management Computer System programs to broaden their views
in core areas of computer sciences, in addition to Programming Languages and Data Struc
tures.
Lastly, this course is needed to
take advantages of the faculty expertise in the department
and
give students background in an active area of undergraduate research
. Two faculty
members
are
actively doing research in
the
Artificial Intelligence fiel
d.
Relationship to program assessment objectives:
The
A
ssessment
C
ommittee in the
D
epartment of Mathematical and Computer Sciences
has
identified
5 universal objectives that address cognitive process
es
involved in the learning. They
are Analytical Reason
ing, Conceptual/Foundational Understanding, Pattern Recognition,
Problem Solving, and Synthesis. Adding the course “Introduction to Artificial Intelligence” will
help strengthen all of these assessment objectives. Specifically, knowledge
representation
a
nd
reasoning approaches are two
of the most important topics addressed in this course. Students will
have a chance to learn about first
-
order logic, rule
-
based reasoning based on first
-
order logic,
Bayesian networks, and reasoning under uncertainty. This cou
rse also provides students with
concept
ual understanding and representation of the world through
finite state machines. One of
the most important topic
s
in Artificial Inte
lligence is machine learning
. The understanding of
different learning approaches (e.g
.
learning by observations, supervise
d
learning, reinforcement
learning, unsupervised learning and so forth) will directly contribute to reinforce the students’
ability to recognize when a particular methodology is appropriate. Problem solving
is another
i
mportant topic in Artificial Intelligence
that helps students define the problems (e.g
.
search,
modeling users)
and solve the problems using Artificial Intelligence
techniques. Lastly, this
course
brings
the theory and empirical research together to hel
p s
tudents understand and use
Artificial Intelligence
techniques and programming tools to solve real world problems.
Budgetary impact:
Staffing:
The p
rimary instructor
is Dr. Hien Nguyen,
Assistant Professor in Computer Sciences
.
Dr. Nguyen is currently tea
ching both M
anagement Computer Systems
and
Computer Science
courses.
There is also another faculty in the department
,
Dr.
Lopa Mukerjee, whose background
is also Artificial Intelligence. This course will likely become a course offered every year
or
every
s
emester,
allowing bo
th instructors to participate and use
their research skills and interests to
convey to students.
Additional fund may be needed to cover overload if the situation arises.
Academic unit library and service & supply budget
:
A new book wi
ll likely be ordered by the library for the proposed course. This book
Artificial
Intelligence: A Modern Approach, 2/E costs $123. The course is low budget which requires only
a small amount to funding to sustain.
Impact on campus instructional resource u
nits:
T
his course will likely enter into a rotation
with existing upper le
vel courses in Computer Science
.
If the proposal of the new gaming major
is approved, this course will serve as one
of the
elective
course
s
in the technology thread for this
major.
Revised 10/02
3
of
11
Laboratory facilities:
adequate lecture space is available. This course can be taught in any
lecture rooms.
Course description:
This course introduce
s
basic
artificial intelligence
principles
including
simple representation
schemes, problem solving parad
igms, constraint propagation,
search strategies,
and
learning
approaches.
K
nowledge representation, natural language processing, gaming, machine learning
and user modeling will be explored.
S
tudents
should
have written
moderately complex
computer
programs
in a high level language
.
Course objectives and tentative course syllabus:
Course information,
objectives
, weekly description, project description and grade information
(page
4
-
9
), and bibliography (page
1
0
)
.
Revised 10/02
4
of
11
COMPSCI 370
: Introduction to Artificial
Intelligence
Required Texts
:
Artificial Intelligence: A Modern Approach (2nd Edition) (Prentice Hall Series in Artificial
Intelligence). 2002. ISBN: 0137903952
Course Description
This course will introduce the basic principles in artificial intelligence
. It covers simple
representation schemes, problem solving paradigms, constraint propagation, search strategies,
and learning approaches. Areas of application such as knowledge representation, natural
language processing, gaming, machine learning and user
modeling will be explored. This course
is intended for students who have at least one high
-
level language and have written computer
programs of at least moderate complexity.
Prerequisite: MCS 220
or equivalent course with C++
(
new COMPSCI
202 in the Progr
amming
thread of new gaming major)
Course
Objectives
1
Given a
basic Artificial Intelligence(AI)
problem such as search, gaming, planning,
ma
chine learning
, understand the
theory, and implement algorithms
being used to
solve this problem.
2
Given a real
world problem, be able to identify the parts in which AI techniques can
be applied.
Tentative
Course
Schedule
WEEK
READINGS
Assignment
Description
Project/Exams
1
Part 1, 1.1 to
1.4
Introduction to AI
The focuses are the i
ntroduction to AI
,
highlig
ht the inter
-
disciplinary nature of AI
(relations with other fields such as
psychology, mathematics, sociology), review
the traditional AI and modern AI approaches
and subfields. Lastly, we will learn how AI is
used in software applications.
Homework 1
2
Part 1, 2.1 to
2.4
Intelligence Agents
The focuses are the representation of a
gent
environment
and a
gent behaviors
; s
tructure of
intelligent
agents
; communication among
agents. We discuss the design and
implementation of an automated, intelligent
agent.
Pr
oject 1 (agent
-
behaviors and
search problems)
Revised 10/02
5
of
11
3
Part 2, (3.1,
3.3, 3.4,
3.6;4.1, 4.2)
Problem solving by Searching
The focuses are the p
roblem solving agents
;
the use of searching algorithms for solutions;
algorithm and complexity analyses of
uninformed s
earches and informed searches
including depth first search, breadth first
search, A* search.
Homework 2
4
Part 2 (4.5)
Online search
The focuses are the description of online
search; the challenges faced by development
of algorithms for online search; an
d the use of
intelligent agents to develop personalized,
online search.
Project 1 is due.
5
Part 2 (6.1
-
6.8)
AI for games
The focuses are the optimal decision in game
development; imperfect, real
-
time decision.
Specifically, t
he minimax algorithm
,
alpha
-
beta pruning
technique will be discussed. At
the end, evaluation functions and cut
-
off
search as well as s
tate of the art game
programs
are studied.
Homework 3
Project 2
(AI game with
finite state
machine)
6
Finite State machine
The focuses include the p
resentation on what
finite state machine is, how to implement
finite state machine for game development.
Students will study a specific example of
implement a simple game using finite state
machine.
Homework 4
7
Part 3 (7, 8,9)
Logical Agents
The focuses
include the introduction of f
irst
order logic
and i
nference on first order logic
;
representation of a knowledge base using first
order logic in a specific application.
Homework 5
8
Part 3 (10)
Knowledge Representation
The focuses are the representation
and
reasoning techniques of basic knowledge
representation schemes (rule
-
based, action
-
based, and situation
-
based), discussion of
online applications using those knowledge
representation techniques.
Project 3
Bayesian
networks
9
Part 3 (10)
Knowledge Rep
resentation
The focuses are the construction and
representation of ontology; use of ontology in
real
-
world applications.
Midterm Exam
Homework 6
10
Part 5 (14,15)
Bayesian Networks
Homework 7
Revised 10/02
6
of
11
The focuses are to create, update Bayesian
networks for a real
-
world ap
plication, set
evidences, perform belief update, and perform
belief revision.
11
Part 6 (18, 19)
Machine Learning
The focuses are to create decision tree from
observations, choose attribute sets, assess
performance of a learning algorithm, and
size
of training set.
Project 4
(Machine learning
techniques)
12
Part 6 (20)
Statistical Learning Methods
The focuses are to understand f
undamental
learning models including naïve Bayes
model, learning with hidden Markov model,
nearest
-
neighbor model;
apply these models
to solve a real world problem.
Homework 8
13
Part 7
Natural language processing
The discussion and presentation focus on
probabilistic context
-
free grammars,
components of a natural language processing
system, introduction of a typical
information
retrieval system, boolean model and vector
space model for retrieval.
Homework 9
14
Elective topics
User modeling
The study focuses on applying Bayesian
networks and Machine learning techniques to
build a user model; ty
pes of user models
(dyn
amic and
static), and evaluations of the
effectiveness of a user model.
Homework 10
15
Elective topics
Designing game in AI
Using “learning
-
by
-
doing” approach, we
introduce students to develop a game with
embedded AI techniques which focus on real
-
time st
rategy and t
urn
-
based strateg
y
16
Exam week
Final exam
TBA
Project
Description
Project 1: Search and intelligent agents
Objectives:
(i)
Understand and implement the fundamental search algorithms in artificial intelligence
(ii)
Understand and apply the testin
g technique to evaluate these algorithms
(iii)
Understand the framework of an intelligent agent
(iv)
Be able to implement an intelligent agent to solve a small problem using a search
algorithm.
(v)
Be able to create a test for the implemented agent.
Revised 10/02
7
of
11
Content:
Part 1: Imp
lement and test a set of fundamental search algorithms in artificial intelligence
including:
Breadth First Search
Depth First Search
A* search
Part 2: Implement an intelligent agent that solves a real world problem (such as placing
cameras in a building)
using one of these search algorithms.
Any programming language can be used to implement this project
Evaluation:
This project will be evaluated on the following criteria
(i)
Accuracy of the implemented search algorithms
(ii)
Accuracy of the implemented agent
(iii)
Test
cases for search algorithms and
agent.
(iv)
Timeliness of the submission
Project 2
: Artificial Intelligence for games
Objectives
:
(i)
Understand how
artificial intelligence techniques can be used to develop a game
(representation of the game domain, knowledge ba
se, and user interface)
(ii)
Be able to create and implement a finite state machine
(iii)
Be able to integrate different techniques for representing the knowledge as well as
interfacing with players.
Content
Simple soccer game
Java
or C++
is recommended to be use
d to implement this project
Assessment:
This project will be evaluated on the following criteria
(i)
Accuracy of finite state machine
(ii)
Accuracy of the implemented actions
(iii)
Test case for the team players.
(iv)
Timeliness of the submission
Bonus: User friendly inter
face.
Project 3: Bayesian networks
Objectives
(i)
Understand how to represent the uncertainty in a real world domain
(ii)
Understand how to create a Bayesian network, set up evidence and do belief update as
well as belief revision
(iii)
Learn how to predict
using Bayes
ian networks
Revised 10/02
8
of
11
Content
Using Genie to create Bayesian networks for a real world domain (e.g stock forcast, game
prediction). Do inference over the constructed Bayesian networks and evaluate its accuracy.
Assessment:
This project will be evaluated on the f
ollowing criteria
(i)
Accuracy of the creation of the Bayesian network
(ii)
Acc
uracy of the prediction
(iii)
Timeliness of the submission
(iv)
Readability of the report.
Project 4: Machine learning
Objectives:
(i)
Understand how to use machine learning techniques to learn from
observed data and
predict
(ii)
Apply these techniques to solve learning problems in real world.
Content
Given data in a popular domain such as image classification, prediction of movie/book
preferences, u
se
Weka software to learn the rules, and classified newl
y observed data.
Assessment
This project will be evaluated on the following criteria
(i)
Acc
uracy of the test files
(ii)
Timeliness of the submission
(iii)
Readability of the report.
Grading
Policy
GRADABLE
Percentage
4 Projects
50%
10 Homeworks
25
%
Midterm exam
10
%
Final exam
15%
Total
100%
Letter
Grade
Percentage
Letter
Grade
Percentage
A
94 to 100%
A
-
90 to 93%
B+
87 to 89%
B
84 to 86%
B
-
80 to 83%
C+
77 to 79%
C
74 to 76%
C
-
70 to 73%
D+
67 to 69%
D
64 to 66%
D
-
60 to 63%
F
Less than 60%
Revised 10/02
9
of
11
Technology
requirement
These applications are used in this class:
Either
Java Development Kit (JDK):
http://java.sun.com/javase/downloads/index.jsp
Or
Visual C++ from Microsoft.
Weka (machine learning): h
ttp://www.cs.waikato.ac.nz/ml/weka/
Genie/Smile (free download for Bayesian networks) from
http://genie.sis.pitt.edu/
Revised 10/02
10
of
11
Bibliography:
(Key or essential references only. Normally the bibliography should be no more
than one or two pages in length.)
Antoni Ligeza.
Logical Foundations for Rule
-
Based Systems (Studies in Computational
Intelligence).
Springer; 2nd edition. 2006.
Finn V. Jensen
, Thomas D. Nielsen.
Bayesian Networks and Decision Graphs
. Information
Science
and Statistics. 2007
Ian H. Witten, Eibe Frank.
Data Mining: Practical Machine Learning Tools and Techniques.
Second Edition. Morgan Kaufmann Series in Data Management Systems. 2005.
Ian Millington.
Artificial Intelligence for Games
. The Morgan Kaufmann S
eries in Interactive 3D
Technology. 2006.
Judea Pearl.
Causality: Models, Reasoning, and Inference
.
Cambridge University Press. 2000
Lin Padgham, Michael Winikoff.
Developing Intelligent Agent Systems: A Practical Guide.
W
iley Series in Agent Technology
.
2004.
Max Chevalier, Christine Julien, Chantal Soule
-
Dupuy.
Collaborative and Social Information
Retrieval and Access: Techniques for Improved User Modeling
. IGI Global. 2009
Paul R. Cohen.
Empirical methods for artificial intelligence
. MIT Press. 1995
Pe
ter Allen.
Natural Language Understanding
. Addison Wesley; 2 edition August 13, 1994.
Ralph L. Keeney, Howard Raiffa.
Decisions with Multiple Objectives: Preferences and Value
Trade
-
Offs.
Cambridge University Press. 2008.
Ricardo Baeza
-
Yates, Berthier Ribe
iro
-
Neto.
Modern Information Retrieval
. Addison Wesley; 1st
edition. 1999.
Ronald Brachman,
Hector Levesque.
Knowledge Representation and Reasoning
. The Morgan
Kaufmann Series in Artificial Intelligence. 2004.
St
uart Russell and Peter Norvig,
Artificial Intelligence: A Modern Approach
.
2nd Ed. Prentice
Hall, 200
2
.
Thomas Mitchell.
Machine Learning
. Mcgraw
-
Hill International Edit. 1997.
Notes:
1.
Contact the Registrar's Office (x1570) for available course numbers. A list of subject areas can be
found at
http://acadaff.uww.edu
\
Handbook
\
SubjectAreas.htm
2.
The 15 and 25 character abbreviations may be edited for consistency and clarity.
Revised 10/02
11
of
11
3.
Please submit electronically when approved at the
college level
-
signature sheet to follow in hard copy.
Enter the password to open this PDF file:
File name:
-
File size:
-
Title:
-
Author:
-
Subject:
-
Keywords:
-
Creation Date:
-
Modification Date:
-
Creator:
-
PDF Producer:
-
PDF Version:
-
Page Count:
-
Preparing document for printing…
0%
Σχόλια 0
Συνδεθείτε για να κοινοποιήσετε σχόλιο