National Computing Education Accreditation Council

NCEAC

NCEAC.FORM.001-D

COURSE DESCRIPTION FORM

INSTITUTION National University of Computer and Emerging Sciences (NUCES-FAST)

BS(SE, CS)

PROGRAM (S) TO BE
EVALUATED

A. Course Description

1 NCEAC.FORM.001.D
 National Computing Education Accreditation Council
NCEAC

NCEAC.FORM.001-D

Course Code xx
Course Title Deep Learning for Perception
Credit Hours 3
Prerequisites by Nil or Machine Learning
Course(s) and Topics
Assessment Mid-I: 12.5
Instruments with Mid-II: 12.5
Weights (homework, Assignments: 15
quizzes, midterms,
Project: 10
final, programming
assignments, lab work, Final: 50
etc.)
Course Coordinator Dr. Jawwad A Shamsi
URL (if any)
Current Catalog Introduction to Machine Learning and Deep Learning. Supervised Learning, Unsupervised
Description Learning, Regression vs classification, Clustering Reinforcement Learning. Bayesian
machine learning models, Ensemble methods, Random forest, Introduction to Neural
Networks, Why Neural Networks, Shallow vs Deep Neural Networks, Neurons, Perceptron
learning algorithm, weights, bias and variance, Training / Testing datasets, Activation
functions sigmoid, RELU, tanh, layers, input layers, dense layers, softmax layers, max
pooling, average pooling hidden layers, output layers, Loss functions, Practical aspects of
deep learning, Vanishing Gradient , Batch normalization, stride, pooling, Optimization,
and programming frameworks (Keras and Pytorch), Structuring Deep learning projects,
practical examples and case studies, Convolutional neural networks, Computer vision
examples, Resnets, Skip connections, Ensemble Learning, Inception net, Transfer
learning, Object detection, anchor boxes, Region proposal Networks, IoU, YOLO,
Sequence models, LSTM, Time series prediction, GRU, Attention mechanism, Machine
Translation, Auto-Encoders, Variational Auto Encoders, Generative Adversial Networks
(GANs), Deep learning: Hype and reality, small sample size, Explainable AI, Other issues
and future prospects
Textbook (or I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, MIT Press, 2016
Laboratory Manual
for Laboratory
Courses)
Reference Material 1) Yaser S. Abu Mustafa and Malik Magdoon Ismail, Learning from Data
2) Dive into Deep Learning.

2 NCEAC.FORM.001.D
 National Computing Education Accreditation Council
NCEAC

NCEAC.FORM.001-D

Course Goals
A. Course Learning Outcomes (CLOs)
1. To impart fundamental concepts about Deep Learning and Machine Learning
2. To study and apply various applications and case studies of deep learning in
various domains.
3. To comprehend architecture and limitations of well-established Deep neural
networks DNNs.
4. To develop hands-on skills on well-established deep neural network architectures
and deep learning software frameworks
5. To study directions for future research in deep learning

B. Program Learning Outcomes

For each attribute below, indicate whether this attribute is covered in this course
or not. Leave the cell blank if the enablement is little or non-existent.

1. Academic To prepare graduates as computing professionals 

Education:
2. Knowledge for Apply knowledge of computing fundamentals, 
Solving knowledge of a computing specialization, and
Computing mathematics, science, and domain knowledge
Problems: appropriate for the computing specialization to the
abstraction and conceptualization of computing
models from defined problems and requirements.

3. Problem Identify, formulate, research literature, and solve

Analysis: complex computing problems reaching
substantiated conclusions using fundamental
principles of mathematics, computing sciences, and
relevant domain disciplines.

4. Design/ Design and evaluate solutions for complex 

Development of computing problems, and design and evaluate
Solutions: systems, components, or processes that meet
specified needs with appropriate consideration for
public health and safety, cultural, societal, and
environmental considerations.

5. Modern Tool Create, select, adapt and apply appropriate 

Usage: techniques, resources, and modern computing
tools to complex computing activities, with an
understanding of the limitations.

6. Individual and Function effectively as an individual and as a 

Team Work: member or leader in diverse teams and in multi-
disciplinary settings.

3 NCEAC.FORM.001.D
 National Computing Education Accreditation Council
NCEAC

NCEAC.FORM.001-D

Communicate effectively with the computing 

7. community and with society at large about
Communication: complex computing activities by being able to
comprehend and write effective reports, design
documentation, make effective presentations, and
give and understand clear instructions.

8. Computing Understand and assess societal, health, safety, 

Professionalism legal, and cultural issues within local and global
and Society: contexts, and the consequential responsibilities
relevant to professional computing practice.

9. Ethics: Understand and commit to professional ethics, 

responsibilities, and norms of professional
computing practice.

10. Life-long Recognize the need, and have the ability, to 

Learning: engage in independent learning for continual
development as a computing professional.

C. Relation between CLOs and PLOs

(CLO: Course Learning Outcome, PLOs: Program Learning Outcomes)
PLOs

1 2 3 4 5 6 7 8 9 10

1  
2      
CLOs

3       
4       
5       

Topics Covered in 1. Topics to be covered:

the Course, with
Number of Lectures No. of Contact
List of Topics
on Each Topic Weeks Hours CLO
(assume 15-week Introduction to Machine Learning and 2 6 1
instruction and one-hour Deep Learning. Supervised Learning,
lectures) Unsupervised Learning, Regression
vs classification, Clustering, Bayesian
Machine learning models, Data
distribution Ensemble methods,
Feature engineering, Performance
metrics, Accuracy, Sensitivity,

4 NCEAC.FORM.001.D
 National Computing Education Accreditation Council
NCEAC

NCEAC.FORM.001-D

Specificity, F1-measure, AUC, ROC,

Hands-on-experience

Introduction to Neural Networks, Why

Neural Networks, Shallow vs Deep
Neural Networks, Neurons,
Perceptron learning algorithm,
weights, bias and variance, Training /
Testing datasets.

Activation functions sigmoid, RELU,

tanh, layers, input layers, dense
layers, softmax layers, max pooling,
3 9 1,2
average pooling hidden layers, output
layers, Loss functions, Differentiation,
over and Underfitting models

Forward propagation, Backward

propagation, Vectorization,
Stochastic Gradient descent,
Parameter and Hyperparameter
tunning. Hands-on- programming
exercises
========== MID 1 ==========
Practical aspects of deep learning,
Gradient vanishing, Batch
normalization, stride, pooling
Optimization, and programming
frameworks (Keras and Pytorch) 2 6 1,2,3,4

Structuring Deep learning projects,

practical examples and case studies
Hands on programming exercise
Convolutional neural networks,
Computer vision examples, Face
recognition, facial expression
recognition, edge detection, Feature
visualization.
2 6 2,3,4
Resnets, Inception net, Alexnet,
VGG, Transfer learning, 1D
Convnets, Hands on programming
exercises
Object detection, anchor boxes,
1 3 2, 3, 4
Region proposal Networks, YOLO
========== MID 2 ==========
Sequence models, LSTM, Time 2 6 2,3,4
series prediction, prediction, prophet
facebook model for COVID-19
forecasting, Transformers and Auto-
Encoders
5 NCEAC.FORM.001.D
 National Computing Education Accreditation Council
NCEAC

NCEAC.FORM.001-D

Generative Adversial Networks

(GANs), Attention mechanism

Reinforcement Learning 1 3 4,5

Project Presentations 1 3 1,2,3,4,5,8

Total 16 48 1,2,3,4,5

Laboratory Hands on experience / Tutorials on implementation of deep learning models

Projects/Experiments
Done in the Course
Programming We will have programming exercises in alternate weeks
Assignments Done in
the Course
Class Time Spent on Theory Problem Analysis Solution Design Social and Ethical
(in credit hours) Issues
35 30 30 5
Oral and Written Every student / group is required to design a project using deep leaning models. The
Communications assessment of the project will be based on innovation, practicality, and clarity. Each
group will demonstrate the project to the instructors and peers, deliver a 10 mins project
presentation and a well written project report.

Instructor Name: Dr. Jawwad A Shamsi

Instructor Signature: ____________________________
Date February 1st, 2022

6 NCEAC.FORM.001.D