CS 375 Artificial Intelligence
Syllabus
(Fall, 2004)
Professor: Dr. Joseph T.
Wunderlich
Office: Nicarry 244
Phone: 361-1295
Email: wunderjt@etown.edu
Office Hours: http://users.etown.edu/w/wunderjt/schedules/Schedule
Card f04 joe w.htm
Objectives: Machine Intelligence is
found in many modern-day technologies and can be defined as encompassing all of
the developments in both symbolic artificial intelligence and artificial neural
networks. Traditional symbolic AI uses programmed heuristics and forms of
knowledge representation to produce results in a seemingly more intelligent way
than typical computer programs. Artificial neural networks are a form of
connectionist computer architecture where many simple computational nodes are
connected in an architecture similar to that of biological brains for the
purpose of solving problems which require rapid adaptation or solutions where
underlying governing equations are not known or cannot be easily computed This
course begins with a comparison of human and machine intelligence followed by a
comprehensive and in-depth analysis of current neural network theory and
applications. A study of available neural network computer hardware and
software is included. Ethical issues concerning artificial intelligence are
also discussed. Several mobile robot and robotic arm concepts will be
introduced.
Course Credit: Four
Prerequisites:
·
Computer
Science I (CS 121) (mandatory)
·
Computer
Science II (CS 122) (recommended)
·
Algorithms
and Data Structures (CS 221) (recommended)
·
Calculus
I (Math 121or 117) (mandatory)
·
Calculus
II (Math 122) (mandatory)
·
Linear
Algebra (Math 201) (recommended)
Prerequisite Topics:
·
Derivation
of algorithms (mandatory)
·
Differentiation (mandatory)
·
Integration (mandatory)
·
Calculus
of trigonometric, exponential, and logarithmic functions (recommended)
·
Matrix
manipulation (recommended)
·
Microsoft
PowerPoint (for oral presentations)
·
Proper
documentation of research
Course Text:
·
S.
Haykin, "Neural Networks, A Comprehensive foundation." 2nd ed.
Upper saddle River, NJ: Prentice-Hall, 1999. (ISBN: 0132733501)
Supplimental
·
Introduction
to the theory of neural computation, Hertz, John, 1991
·
Neural
and concurrent real-time systems: the sixth generation, Soucek,
Branko, 1989
·
Neural
and massively parallel computers: the sixth generation, Soucek,
Branko, 1988
Other Recommended
·
Artificial
intelligence: structures and strategies for complex problem solving, Luger, George F, 1998
·
Cambrian
intelligence: the early history of the new AI, Brooks, Rodney Allen, 1999
·
The
human mind according to artificial intelligence, Wagman,
Morton, 1999
·
International
Conference on Robotics and Automation [videorecording],
2000
·
Introduction
to AI robotics, Murphy, Robin, 2000
·
Layered
learning in multiagent systems: a winning approach to
robotic soccer, Stone, Peter, 2000
·
Neurocomputing: foundations of research,
·
Neurotechnology for biomimetic
robots, Ayers, Joseph, 2002
·
Proceedings
/ IEEE International Conference on Robotics and Automation, 1986
·
Proceedings
2000 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2000
·
Pulsed
neural networks, Mass, Wolfgang, 2001
·
Robo sapiens: evolution of a new species, Menzel, Peter, 2000
Grading:
·
Homeworks =8%
·
Project
#1 =20%
·
Project
#2 =22%
·
Midterm
exam(s) =20%
·
Comprehensive
final exam =30%
COURSE GRADE:
(60-62)=D-, (63-67)=D,
(68-69)=D+, (70-72)=C-, (73-77)=C, (78-79)=C+, (80-82)=B-, (83-87)=B,
(88-89)=B+, (90-92)=A-, (93-100)=A
(with any fractional part
rounded to the nearest integer)
Attendance: Exams will primarily cover material
presented in lecture -- some of which may not be found in the texts.
Academic Honesty:
Course Outline:
I.
Neural
networks vs. symbolic artificial intelligence
(1) “Bottom-up” brain models
(2) “Top-Down” brain models
(3) Evolution of neural
networks
i.
Perceptrons
ii.
Learning
rules
iii.
Underlying
neural network mathematical theories
1.
Multivariable
calculus review
2.
Linear
algebra review
iv.
Neural
network applications
v.
Neural
network hardware and software
(4) Symbolic artificial
intelligence
i.
Predicate
calculus
ii.
Knowledge
representation
iii.
A.I.
programming languages
1.
Prolog
2.
LISP
iv.
Expert
systems
II.
Biological
vs. machine intelligence. The following “Mental Abilities” will be discussed by
answering five simple questions:
(1) What can humans do?
(2) What can a simple insect
do? (e.g., a spider)
(3) What can a conventional
computer do?
(4) What can symbolic A.I.
programming do?
(5) What can artificial
neural networks do?
Basic Animal Abilities: Acquire and retain
knowledge, Solve problems, Motor coordination, Acquire energy, Protect self,
Sensory processing, Real-time thought React instinctively, Anticipate, Predict,
Communicate, Generalize, Associate, Recognition patterns, Robust under partial
failure, Autonomous thought, Drive to reproduce, Stability, repeatability,
predictability, Multitask
Complex Abilities: Abstraction, Intuition,
Common sense, Manipulate tools, Heuristics, Inference, Hypothesis testing,
Self-discipline, impulse-control, Ethical behavior, Selective awareness, Open
to inspection, Emotions, Imagination, Creativity, Passion, Playfulness,
Empathy, Leadership, Self-awareness, Awareness of mortality, Group psychology
III.
Merging
neural networks with symbolic artificial intelligence
IV.
Introduction
to robotics
(1) Mobile robots
i.
Path-planning
and obstacle avoidance
ii.
Environmental
mapping
(2) Robotic arms
i.
Kinematics
ii.
Path-planning
and obstacle avoidance
V.
Artificial
humanoids
(1) Emulating human
physiology
i.
Biomechanics
ii.
Senses
iii.
Control
systems
VI.
Ethical
Issues involving machine intelligence
(1) Replacing humans
(2) Aiding humans
(3) Military uses
Assignment: Project #1
Grading: 20% of total course grade
Due
Date:
Late
Penalties: Yes
Last
Revised:
Pretend
you have just been hired by NASA to design an autonomous mobile robot to live
on one of the moons of Jupiter where insect life has just been discovered.
Assume this moon is terrestrial, has no other life, is always cloudy, has nights that are -100 degrees Fahrenheit and days that
are 50 degrees. Also assume one of the other NASA research groups has designed
a digestive system for your robot so that it can derive its energy by consuming
insects that it has caught. Your robots primary goal is to gather data about
its immediate environment (including obstacles, prey, and possibly hostile
insects fighting for territory or food)
Your
specific tasks are:
1.
Research
the behavior of a specific type of earth spider; compare all of its mental
abilities to those listed on the “mental ability matrix” defined in class.
2.
Write
a computer simulation of your robot spider living in its environment. Your user
interface should be as visual as possible; and the robots behavior should be as
complex as possible.
Form
groups of two or three people.
On the project due-date, both a brief written report and a demonstration are due. The
written report must contain:
1) A one page write-up including:
a) background
research findings
b) path-planning strategy
2) All
code (well commented)
Assignment: Project #2
Grading: PROPOSAL: 2% of total course
grade
Due
Date:
Late
Penalties: Yes
Grading: FINAL PRESENTATION AND
REPORT: 20% of total course
grade; letter graded
Due
Date:
Late
Penalties: Yes
Last
Revised:
The
oral proposal should take approximately two to five minutes (not including
answering audience questions). Groups of two or three people are ok; however
all people must speak. The project is a major research project; however you may
also do a design-build if
you’re confident you can complete it. You must print copies of your proposal
presentation and hand it in just before you present.
On
the project due-date, both written and oral reports are due. The oral report must be done using either PowerPoint or
a web page created by you for
the project. It should take approximately ten to fifteen minutes and
contain an appropriate number of visuals. All group members must speak. The
written report must adhere the formatting as shown in “SAMPLE paper 1.doc” in the handouts
folder (i.e., single-spaced, 10-point font, etc.); and should contain:
1) A one to two paragraph
abstract
2) Miscellaneous discussions of details
3) Conclusions
4) A bibliography of referenced material
(must cite at least one peer-reviewed
article)
5) An appendix containing schematics, manufacturers literature, etc.
============================================================================================================
Assignment: Homework #1
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
See
if you can add a couple of abilities to those listed below:
Basic Animal Abilities: Acquire and retain
knowledge, Solve problems, Motor coordination, Acquire energy, Protect self,
Sensory processing, Real-time thought React instinctively, Anticipate, Predict,
Communicate, Generalize, Associate, Recognition patterns, Robust under partial
failure, Autonomous thought, Drive to reproduce, Stability, repeatability,
predictability, Multitask
Complex Abilities: Abstraction, Intuition,
Common sense, Manipulate tools, Heuristics, Inference, Hypothesis testing,
Self-discipline, impulse-control, Ethical behavior, Selective awareness, Open
to inspection, Emotions, Imagination, Creativity, Passion, Playfulness,
Empathy, Leadership, Self-awareness, Awareness of mortality, Group psychology
Assignment: Homework #2
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
Attend
one of the following:
1) IEEE/NASA/WUNDERBOT talks
or
2) IEEE computer society president talk in student center event
space Friday,
or
3) At least 1-1/2 hours of the “Information Technology
Conference”
in Nicarry on Saturday,
Then
write a one page single-space summary of whichever you attended (to be
handed-in in class at the beginning of the first class meeting of the following
week. Also, as an added bonus, will not have class on Monday
or Tuesday during the week of these events.
Note:
I will definitely be at all of (1) and (2) above -- and possibly parts of (3).
Assignment: Homework #3
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
•
Go
to http://www.station1.net/DouglasJones/drake.htm
•
Do
the following:
1.
Read
the rationale for all of the variables in the drake equation
2.
Calculate
a value of N that you believe in
3.
Be
ready to continue our discussion on this
•
(including
how you feel about the Drake Equation)
Assignment: Homework #4
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
1.
Propose
a new version of the survival matrix given out in class. Then apply it to:
a.
A
human
b.
A
spider
c.
A
PC
d.
“Asimo” the HONDA corporation humanoid robot (research this)
Assignment: Homework #5
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
1.
Select
one of the 6 human senses discussed in class. Then:
a.
Research
the biology/physiology of this sense in humans
b.
Research
artificial devices to enhance or substitute for this sense
Assignment: Homework #6
Grading: (part of “assignments” grade)
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
Calculus problems handed out in class
Assignment: Homework #7
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
Select one of the robots demonstrated in the
video shown in class” 1996 IEEE Robotics and automation video proceedings”and find the
corresponding paper in the written proceedings on reserve in the library. Then
summarize an interesting detail that you discover, write about it (one page
single space); then be prepared to tell the class what you found.
Or
Search the above proceedings for a paper that
you consider is about adding AI into a robot and write about it (one page
single space); then be prepared to tell the class what you found.
Assignment: Homework #8
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
Attend Dr. Wunderlich’s
ENGR 100 lecture on Engineering Design in Gibble
auditorium. Then meet a freshman and get to know them. Type a one page summary
of your interaction with this student and include how you have helped them in
any way (this can include long-term planning help – course selection, career
planning, etc.)
Assignment: Homework #9
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
Attend the field trip to the JLG company in
Type a one page summary of your ideas for
possible automation (and potential AI applications) for their facilities.
Assignment: Homework #10
Professor: Dr. Wunderlich
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
1) Run the NN matlab example
“NN2.m” for XOR, AND, and OR, and find the largest learning rate possible for
each. Plot results and hand-in results.
2) Form groups of two or three people. Run the Matlab Neural Network example “Appcr1” for character
recognition. Create a report explaining exercise (including graphs, code, etc.)
See extra credit option on handout.
Assignment: Homework #11
Professor: Dr. Wunderlich
Due
Date:
Late
Penalties: Yes
Last
Revised: ------
Create a
symbolic-AI And/Or graph for the Expert System: “Case
study #1 Doctor’s Office” handed out in class
Friday,