Course
hashtag #501422NN
Announcements,
MidTerm Marks,
Answer Model
Read more announcements here!
Student Feedback and Appreciation of Lectures (download
Excel file)
An artificial neural network, often just
called a neural network, is a mathematical model inspired by biological
neural networks. A neural network consists of an interconnected group of
artificial neurons, and it processes information using a connectionist
approach to computation. In most cases a neural network is an adaptive
system that changes its structure during a learning phase. Neural networks
are used to model complex relationships between inputs and outputs or to
find patterns in data.. (See Wikipedia)
~Course
Information:
Course syllabus 
student version
Code

5014223

Title

Neural Networks

Credit Hrs.

30:3

Semester

2

Academic Year

20122013

Pre/Co Requisites

25013613
Artificial Intelligence

Instructor

Office Hours/ Location

Email

Teaching Assistants

Dr. Salha M. Alzahrani

Mon. 8am12pm Tues. 8am12pm Garwa deanship block, Room 1630

admin@c2learn.com
s.zahrani@tu.edu.sa
Please send your enquiries to one email address not
both!

Mrs. Samsad Mrs. Ameera Almas

Course Format

Lecture Time/Venue

Course Website

Lectures: 28/Section
Lab: 28/Section

Section 266 (Sat. 8am10am) @22 Section 212 (Sun. 8am10am)
@25
Section 243 (Sun. 10am12pm)
@21

www.c2learn.com/neural_networks/

~Synopsis:
This is an introductory course to artificial neural
networks (ANNs). The course starts with a descriptive analogy with
biological neural networks. The evolution of the main models of ANNs is
introduced. In particular, McCulloch and Pitt's Model is discussed through
which the basic concepts of ANNs are explained. The perceptron model is
detailed and the linear separability concept is discussed and shown to be
the reason behind the decline of ANNs in the late 70s. Singlelayer versus
multilayer feedforward neural networks are discussed with handson
exercises. Backpropagation model, as a solution to the limitation of
perceptron learning, is introduced and the learning algorithm is discussed
in depth. Additional ANNs models, as the unsupervised SelfOrganizing Maps,
are introduced as well.
~Course
Objectives:
CO1 : Introduce the main fundamental concepts and techniques of
artificial neural network models. CO2 : Investigate the main artificial
neural network models and their applications. .
~Learning
Outcomes:
Upon successful completion of the course,
the students will be able to:
CLO1 : Understand and describe the analogy between
biological and artificial neuron. CLO2 : Understand how early neural
network systems carried out computations. CLO3 : Understand the learning
process of Perceptrons models. CLO4 : Understand the linear separability
of perceptron models and how it restricts its learnability. CLO5 : Use
the backpropagation algorithm to solve the restrictive learning problem in
perceptrons. CLO6 : Use an unsupervised learning to train models based
on selforganized maps. CLO7 : Evaluate the practical considerations in
applying neural networks in real classification problems. .
~Course
Schedule:
Week

Topic

Slides

Assignment(s)/ Lab Exercise(s)

Quiz(zes)

0 26/1/2013

Course syllabus and structure 
Lecture_0 


1 2/2/2013

Introduction to Biological and Artificial Neural Networks (History
and Evolution)

Lecture_1

Discussion
of research works on neural networks. Sample of a paper entitled
“Review of neural network applications in sleep research”, Journal
of Neuroscience Methods, Elsevier.
Introduction
to MATLAB environment and the Neural Networks Toolbox.


2 9/2/2013

The Neuron as a Simple Computing Element: McCulloch–Pitts (MCP)
Neuron

Lecture_2

More
on MATLAB environment and the Neural Networks Toolbox.


3 16/2/2013

Perceptrons I: Linear Separability Concept and
Learning Algorithm

Lecture_3

NN_Handout_Perceptron_AND
NN_Handout_Perceptron_OR
Using
the Neural Networks Toolbox and Wizard with AND, OR, and Irish
datasets.


4 23/2/2013

Handson Exercises of Perceptrons Learning: Logical AND, OR,
NAND
Singlelayer and
Multilayer Feed Forward Neural Networks

Lecture_4

NN_Handout_Perceptron_AND_3Inputs
NN_Handout_Perceptron_OR_3Inputs
NN_Handout_Perceptron_NAND_3Inputs
Exercise(s)
and practical problems on perceptrons.


5 2/3/2013

More
Handson Exercises of Perceptrons Learning

Lecture_5

NN_Handout_Single_Layer_Feedforward
NN_Handout_MultiLayer_Feedforward
(3221)
Exercise(s)
and practical problems on perceptrons.

Quiz
IA
Quiz IB
Quiz IC

6 9/3/2013

Multilayer Neural Networks and MATLAB Demos.

Lecture_6

NN_Handout_MultiLayer_Feedforward
(232)
Exercise(s)
and practical problems on singlelayer versus multilayer
feedforward neural networks.


7 16/3//2013

Mid Semester Exams





 23/3//2013

Mid Semester BREAK





8

Backpropagation Learning Algorithm I

Lecture_7

Exercise(s)
and practical problems on backpropagation neural networks.


9

Backpropagation Learning Algorithm II

Lecture_8

NN_Handout_BP_XOR
NN_Handout_BP_Bipolar_XOR
NN_Handout_BP_ZIP
(.doc files)
Assignment
2
Exercise(s)
and practical problems on backpropagation neural networks.

Solution
of BP_XOR (.doc)

10

Accelerated learning in multilayer neural networks.

Lecture_9

MATLAB
Demos
Exercise(s)
and practical problems on backpropagation neural networks.
Assignment
2 due
L

Quiz II: will be held during your lab

11

Backpropagation Learning Algorithm: Example of a Real
Classification Problems.

Lecture_10 
Exercise(s)
and practical problems on backpropagation neural networks.


12

Unsupervised Learning using SelfOrganizing Maps (SOMs)

 
Exercise(s)
and practical problems on backpropagation neural networks.

Quiz III
A&B

13

Review of the Course



Lab
Revision.


14

Lab Final Examination




15&16

Final Exam





~Course
Assessment:
Quizzes: 10, Assignment(s)/Project(s): 5, Lab: 15,
Mid Term Exam: 20, Final Exam 50.
~Text/References:
Main Textbook: Sivanandam, M.
Introduction to Artificial Neural Networks, Sangam Books Ltd., 2003.
ISBN10: 8125914250.
Other Reference(s): Fausett, L.,
Fundamentals of Neural Networks, Prentice Hall, 1994.
Negnevitsky, M., Neural Networks in Artificial
Intelligence: A Guide to Intelligent Systems, 2nd ed., Addison Wesley,
Harlow, England, 2005.
~Course
Policies:
Attendance

it is compulsory for students to attend at least 75% of the whole
time allocated to the course. More than four (4) unjustified
absences will make you ineligible for the course.

Quizzes

students are informed about the date of their quizzes. Check the
above course schedule to see when a quiz is due.

Assignments

students are expected to submit their assignments in a timely
manner.

Late Submissions

late submission may be subjected to a decrease in the grade that is
proportional to the delayed days.

Midterms and Finals

exams are closed books. Students who did not attend the midterm due
to exceptional situations (e.g., medical emergency) will see their
midterm result calculated from their final. 0.75 * Final exam mark
will be used as their midterm result.

Plagiarism

student caught cheating will be awarded a 0 in the exam and may be
subject to university disciplinary actions.

Disability

students with disability need to approach the course instructor so
that commodities will be put in place to assist them. This is done
in a personal and confidential manner.

