A Course Syllabus

for

CS 5720

Neural Network & Deep Learning

(Three semester hours credit)

in the

School of Computer Science and Mathematics

of the

COLLEGE OF HEALTH, SCIENCE AND TECHNOLOGY

Catalog Description:
This course provides opportunities to learn various deep learning techniques to model
data for classification and prediction. Concepts and methods are illustrated with real-
world applications.

UNIVERSITY OF CENTRAL MISSOURI

Lee Summit's, Missouri

Fall 2024
 Instructor Email Office# Office Hour
Muhammad Zubair Khan mzkhan@ucmo.edu WCM 215 TBA

Course Time and Location

Time: Thursday 3:00 PM 5:40 PM
Location: MIC, LS MO A224
Credits: 3 Credit Hours

Description:
Neural Network & Deep Learning
This class is a Hybrid format. Students will meet in-person at the scheduled time plus one
extra in person day for midterm to be determined by the faculty. For weeks that are not
scheduled as in-person, attendance is required during the scheduled class meeting time via
Zoom. An in-person midterm will be scheduled during the class meeting time of the extra in-
person day. There is a total of 7 in-person days, 7 Zoom days, and one day for in-person
midterm, for 15 meetings. An in-person Final Exam will be given during final exam week.

Personal Zoom Link: https://ucmo.zoom.us/j/5771503689

Technology Requirements
You are expected to have the computing resources necessary to complete this course through
personal, University channels (e.g., remote computer labs) or both. Please contact me if you
will be without email access for a short-term basis during this course. We can make alternate
arrangements should your reason for being without computer access warrant an
accommodation (note: travel for vacation/work does not necessitate accommodations).
Below is a list of some helpful computer requirements for full participation in this class:

The latest version of GitHub Desktop at https://desktop.github.com/

The latest version of Python IDE (Desktop or Web)
A current word processing software
A headset with microphone
A webcam
 Firefox, Chrome, or Safari browser.
A Broadband Internet connection is preferred. Examples of broadband Internet
connection are high-speed DSL or a Cable modem

Participation Policies
This course is not designed to be self-paced. Rather, you are expected to participate in
class activities along with the group. At the same time, I recognize that there might be
times during the semester that you need to complete work in advance in order to meet
other life demands

Course Time Commitment

Courses, both in-person and online, can be varied in their design and expectations for
student involvement and time. If this course were completely taught face-to-face during the
16-week session, one would expect to be in class (engaged in lecture and discussion) for
approximately 3 hours/week and spend additional time outside of class in preparation for
active course engagement and course assessment (reading, completing assignments). So,
you should be prepared to spend similar amounts of time engaged in this course in an
online format. Compared to traditional face-to-face learning environments, you should
anticipate that in this course, you will experience less time in passive learning activities
(i.e., lecture). And more time engaged in active learning activities and communication with
your peers and me (i.e., In-class programming (ICPs), Quizzes via Zoom, and discussion
board postings).

Expectations of Faculty in this Course

We will aim to respond to all email questions within 48 hours. Questions that are posted to
the General Questions Discussion Board will be answered within 24 hours. Weekly
assignments will typically be graded by Monday of the following week. Exams, in general,
will have a two-week grading turn-around timeframe.

Purpose of the Course

The main purpose of this course is to introduce students to deep learning, the study of
computer systems that improve their performance automatically through experience.
Students will learn the latest deep learning algorithms and models that constitute typical
learning systems. They will also gain the necessary foundations and background to both
build practical deep learning systems and conduct research in state-of-the-art problems.

Intended Objectives
After finishing this course, you will have the knowledge to be able to write
Machine Learning and Deep Learning programs
1. Complete the weekly In-class programming assignments (ICPs)
2. Complete weekly quizzes
3. Complete course project
Student Learning Outcomes
Computer Science Outcomes
Apply algorithmic principles and computing theories to solve advanced
problems in Computer Science.
Apply design and development principles in construction of computer- based
systems of varying complexity.
Be able to use current techniques, skills, and tools necessary for computing
practice.
Recognize the need for and engage in continuing professional development.

Cybersecurity Outcomes
An ability to apply algorithmic principles and formal models to solve
advanced problems in cybersecurity and computing.
An ability to communicate effectively to a range of audiences, work
effectively in a team environment, and recognize the need for continual
professional development.

Course Content Outline

A. Textbooks:
Recommended:
Machine Learning and Deep Learning with Python, scikit-learn and
TensorFlow by Sebastian Raschka and Vahid Mirjalili, 2017.
Deep learning with python by francois Chollet, 2017.
Deep learning with keras by Antonio gulli et. al, 2017.
Deep learning by Ian Goodfellow et. Al, MIT Press, 2016.
Learn Python the Hard Way by Zed Shaw, Third Edition

Essential Links:
https://scikit-learn.org/stable/
https://numpy.org/
https://matplotlib.org/
https://pandas.pydata.org/
https://keras.io/
https://www.tensorflow.org/
Lesson Description Tools/Technologies
Introduction and Fundamental concepts. Python /
Week-1
Python Features, Applications Jupyter Notebooks
Python /
Week-2 Data Normalization and Performance Evaluation Jupyter Notebooks
Introduction to supervised learning,
Simple Linear Classifier Python/
Week-3
K-Nearest Neighbor Jupyter Notebooks
Decision Tree
Terminology in machine learning,
Support Vector Machines
Logistic Regression Python/ Jupyter
Week-4 Notebooks/
Naïve Bayes
Numpy/Matplotlib
Perceptron
Neural Networks
Matplotlib/Pandas/NumPy/
Week-5
Data Analysis, Unsupervised learning Jupyter Notebooks
scikit-learn (sklearn) /NumPy/
Week-6
Data Analysis, Dimensionality reduction Jupyter Notebooks

NLTK/Python/
Week-7
Natural Language processing, Python NLTK package, Stemming, POS, Jupyter Notebooks
Lemmatization, N-grams, Name entity recognition
Brief introduction to deep learning, Installation, TensorFlow graphs and Google Colab/
Week-8
sessions Jupyter Notebooks
Constants, Basic operations, Tensor Ranks, Variables, Placeholders and Python, scikit-learn, Keras, NumPy,
Week-9
feeding inputs, Lazy loading, TensorBoard, loss functions. matplotlib, Jupyter Notebooks
Python, scikit-learn, Keras, NumPy,
Week-10
Neural Networks, Transfer Learning matplotlib, Jupyter Notebooks
Artificial Neural Networks, types of artificial neural networks, activation
Python, scikit-learn, Keras, NumPy,
Week-11 function, types of activation functions and Image/Text classification with
matplotlib, Jupyter Notebooks
CNN.
Python, scikit-learn, Keras, NumPy,
Week-12
Logic behind Activation functions, RNN, LSTM, Implementing Text Generator matplotlib, Jupyter Notebooks
Basics and types of Autoencoder. Dimensionality reduction with MNIST data Python, scikit-learn, Keras, NumPy,
Week-13
with Autoencoders. matplotlib, Jupyter Notebooks
Generative Adversarial Networks (GAN s) Python, scikit-learn, Keras, NumPy,
Week-14
matplotlib, Jupyter Notebooks
Procedures/Assessment
The lecture format will be the hybrid mechanism used in the course. Computer
demonstrations in the classroom will be used whenever appropriate.
Programming assignments relating to machine learning and deep learning techniques
will be assigned, specifically building and testing models for various applications. A
brief report should be submitted with each assignment.
d based on their understanding of the subject
material demonstrated through multiple assignments, quizzes, project, a mid-term
and a final exam.

Late Submission Policy on Assignments

Assignments that completed after the due date will no longer be accepted.
1. '-10' points for one day late submission
2. '-20' points for two days late submission
3. No submission will be allowed after Two days unless permission was provided.

DO NOT EMAIL your work, always use the appropriate platform to submit and
your work should be original and independent.
Tasks Contribution % of the Final Grade Point Value
Assignments (10-12) Individual 30% 100 points each
Quizzes (2) Individual 5% 100 points each
Journal club (1) Individual 10% 100 points each
Project (1) Team of (4) 20% 100 points each
Exams (2) Individual 35% 100 points each
Total 100%

Grading
Grading Scale
Percent (%) Grade
90 100 A
80 89 B
70 79 C
60 69 D
0 60 F
Midterm and Final Exam: The exams will consist of questions that emphasize the principles and
analytic techniques for machine learning and deep learning. In addition, the exams will bedivided
into two sets of questions: one set for programming questions and other emphasize on p
concepts and techniques.

Quiz: pop quizzes are given to make sure that students are keeping up with the class and understand
the necessary facts.

Assignments: Students will need to complete the required assignments in time.

1. Once finished your assignment push your source code to your repo and explain the work
through the ReadMe file properly. Make sure you add your student info in the ReadMe file.
2. Comment your code appropriately (show your understanding). IMPORTANT.
3. Make a simple video about 2 to 3 minutes which includes demonstration of your work and
explanation of code snippets.
4. Any submission after provided deadline is considered as a late submission.

Term Project: A student will complete a term project. The project is 25% of your grade. You can
do it either individually or in a group of 4. A good project is one that applies one or more learning
algorithms covered in class, in novel ways to a dataset. An excellent project is a research project that
will result in a paper at a major conference. The project will provide you with a unique opportunity
for exploring one or more areas of deep learning that we did not cover in depth. Some examples are
graphical models, collaborative filtering, inductive logic programming, topic models, language
models, GAN s, and Graphical neural networks. You should choose a data set, apply learning
techniques from these fields to it and compare their performance with the techniques covered in
class.

Project Submission Guidelines:

1. The project submission is with a team of (4) students max.
2. Submit your source code and documentation to GitHub, ReadMe file must have team
information and short introduction.
3. Comment your code appropriately.
4. Video submission (3 to 5 min video showing the demo of the project, with brief voiceover on
the code explanation)
5. Submit the unplagiarized report.
6. Submission after the deadline considered a late submission. (Check the Late Submission Policy
on Assignments in the syllabus)
7. The report should follow the IEEE format and should include the following details:
a. Introduction
b. Objectives
c. Approaches/Methods
d. Workflow
e. Datasets (if applicable)
f. Parameters (if applicable)
g. Evaluation & Discussion (if applicable)
h. Conclusion
Rubric Details

Assignments
Criteria Novice Competent Proficient
ReadMe file Completed Missing Student info. Missing some Meets all the ReadMe
Missing Short information. (>=80 file requirement. (>=90
Description. to <=90) to <=100)
Missing Video link (if
applicable).
(>=70 to <=80)
Video No Video. Video missing Video with all details
proper and proper technical
(0) explanation. explanation.
No Audio. (>=85 to <=100)
(>=50 to <=85)

Completeness of given It is partially solved. Completely solved. It is completely solved

assignment (<=75) (>=75 to <=95) in an efficient way.
(>=90 to <=100)
Commenting the code Not useful comments. Appropriate Appropriate
(<=70) comments. (>=70 to comments. (>=90 to
<=90) <=100)
Time of submission Submission after the It is submitted on the It is submitted on or
due date. Check the deadline. No score will before the deadline.
'Late Submission Policy be deducted from the No score will be
on obtained score. deducted from the
Assignments' section obtained score.
in the syllabus
 Project
Criteria Novice Competent Proficient
ReadMe file Completed Missing Student info. Missing some Meets all the ReadMe
Missing Short information. (>=80 file requirement. (>=90
Description. to <=90) to <=100)
Missing Video link (if
applicable).
(>=70 to <=80)
Video No Video. Video missing Video with all details
proper and proper technical
(0) explanation. explanation.
No Audio. (>=85 to <=100)
(>=50 to <=85)

Report Basic report. (>=80 to Report on the Report with all details
<=85) required details. and making it easy to
(>=85 to <=95) follow and understand
Visually looking good.
(>=90 to <=100)
Report similarity >30 % <30 % <15 %
(>=80 to <=85) (>=85 to <=95) (>=90 to <=100)
Completeness of given It is partially solved. Completely solved. It is completely solved
assignment (<=75) (>=75 to <=95) in an efficient way.
(>=90 to <=100)
Commenting the code Not useful comments. Appropriate Appropriate
(<=70) comments. (>=70 to comments. (>=90 to
<=90) <=100)
Code similarity >50 % <50 % <30 %
(>=50 to <=70) (>=70 to <=90) (>=90 to <=100)
Time of submission Submission after the It is submitted on the It is submitted on or
due date. Check the deadline. No score will before the deadline.
'Late Submission be deducted from the No score will be
Policy on obtained score. deducted from the
Assignments' section obtained score.
in the syllabus
Submission (including Submission with Submission with all Submission with all
feedback) partial details. Miss required details. (>=90 required details. (>=90
any important details. to <=100) to <=100)
(<=90)

Presentation Basic presentation Presentation with Presentation with

(Individual effort) with not explaining explaining the crucial crucial details of the
the crucial details of details of the assignment comparing
the assignment. (>=80 assignment. (>=85 to with a similar project.
to <=85) <=95) (>=90
to <=100)
Americans With Disabilities Act
Students with documented disabilities who are seeking academic accommodations should
contact the Office of Accessibility Services, Union 220, (V) (TTY) 660-543-4421.

Accessibility Services
Students with disabilities who are seeking accommodation should contact the Office of
Accessibility Services at Union 222, 660-543-4421. If you want to share information about
your needs that I should be aware of, such as emergency medical information or special
arrangements for field trips or internships, please see me privately after class or during office
hours.

Sexual Discrimination and Sexual Misconduct Statement

The University of Central Missouri seeks to foster a safe and healthy environment built on
mutual respect and trust. Sex discrimination, including sexual harassment, sexual violence,
and other forms of sexual misconduct will not be tolerated. All faculty and most staff are
considered mandated reports by the University and must disclose all information they
receive about sexual misconduct to the Title IX Coordinator. As a faculty or staff member
of the University, I am a mandated reporter. This means I am required to report information
shared with me regarding sex discrimination and sexual misconduct.

If you, or someone you know, has experienced sex discrimination or sexual misconduct,
please know assistance and options are available. UCM strongly encourages all members of
the community to seek support and report incidents of this nature to the Title IX Coordinator.
Anyone who wishes to report sexual misconduct, to learn more about the University process
and options available, or to utilize a confidential resource, please visit ucmo.edu/titleix.

Diversity, Equity, and Inclusion

The University of Central Missouri strives to develop a campus environment that welcomes
and recognizes all dimensions of diversity and inclusiveness. What this means is that all
students are welcomed in the classroom, and differences are to be recognized rather than
erased or denied. Dimension of diversity can include sex, race, age, national origin,
ethnicity, gender identity and expression, intellectual and physical ability, sexuality, income,
faith, and non-faith perspectives, socio-economic class, primary language, family status,
military experience, and more. Inclusive learning is facilitated by creative and innovative
thought and mutual respect; being in this classroom means that you, your faculty member,
and your peers pledge to foster a welcoming and equitable environment for all.