ECTS - Advanced Deep Learning Techniques and Applications

Advanced Deep Learning Techniques and Applications (CMPE452) Course Detail

Course Name	Course Code	Season	Lecture Hours	Application Hours	Lab Hours	Credit	ECTS
Advanced Deep Learning Techniques and Applications	CMPE452	Area Elective	3	0	0	3	5

Pre-requisite Course(s)
N/A

Course Language	English
Course Type	Elective Courses
Course Level	Bachelor’s Degree (First Cycle)
Mode of Delivery
Learning and Teaching Strategies	Lecture, Question and Answer, Drill and Practice, Problem Solving.
Course Coordinator
Course Lecturer(s)
Course Assistants
Course Objectives	The course objective is to equip students with a good understanding of deep learning principles, enabling them to design, implement, and evaluate advanced neural network models for various real-world applications.
Course Learning Outcomes	The students who succeeded in this course; Grasp advanced concepts of artificial neural networks, including hands-on experience with Python programming language. Develop proficiency in deep supervised learning techniques and backpropagation algorithms, applying them to datasets from platforms like Kaggle. Understand and implement convolutional neural networks for tasks such as object recognition, image segmentation, and vision-based navigation. Explore structural prediction methods and natural language processing applications within deep learning frameworks. Apply optimization techniques to enhance AI model performance, reduce overfitting, and effectively initialize neural networks. Investigate the utilization of deep learning in specialized domains and assess current challenges and future trends in the field.
Course Content	Artificial intelligence, machine learning and deep learning, mathematical building blocks of neural networks, supervised learning, backpropagation, CNNs, object recognition, image segmentation, feature extraction, NLP, optimization techniques.

Weekly Subjects and Releated Preparation Studies

Week	Subjects	Preparation
1	Course Introduction, Introduction to Machine Learning	Course Book – Ch. 1.1
2	Introduction to Neural Networks, Coding in Python (Artificial Brain Development)	Course Book – Ch. 2.1
3	Introduction to Deep Learning	Course Book – Ch. 1.2, Ch. 1.3
4	Deep Supervised Learning I, Deep Supervised Learning II	Lecture Notes
5	Part I: Backpropagation Part II: Experimenting on a Deep Learning Model via a Kaggle Dataset	Course Book – Ch. 2.2, Ch. 4.1, Ch. 4.2, Ch. 5.1
6	Technical Progress of Convolutional Networks, Convolutional Networks for; Multiple Object Recognition, Visual Object Detection, and Simple Object Recognition	Course Book – Ch. 3.1
7	Midterm Exam
8	ConvNet for Segmentation and Vision-Based Navigation, Convolutional Networks in Image Segmentation and Scene Labeling, Convolutional Networks for Real Object Recognition	Course Book – Ch. 3.1
9	ConvNets as Generic Feature Extractors, Image Similarity Matching with Siamese Networks Embedding, Accurate Depth Estimation from Stereo, Body Pose Estimation, Vision Project Ideas, Examples of Deep Learning and Convolutional Networks in Speech, Audio, and Signals, Software Tools and Hardware Acceleration for Convolutional Networks	Course Book – Ch. 3.1, Ch. 3.3
10	Structural Prediction and Natural Language Processing	Course Book – Ch. 8.1, Ch. 8.3
11	Part I: More Backpropagation Part II: Semi-supervised Image Recognition	Course Book – Ch. 6.1, Ch. 6.2
12	Techniques (Optimization, Reducing Overfitting, Initialization)	Course Book – Ch. 5.3, Ch. 9.1, Ch. 9.2
13	Coding in Python (Image Segmentation)	Deep Learning with Python, Second Edition by Francois Chollet – Ch. 9.2
14	Disaster Risk Monitoring Using Satellite Imagery	Course Book – Ch. 10.2, Ch. 10.3
15	Review
16	Final Exam

Sources

Course Book	1. Understanding Deep Learning: Building Machine Learning Systems with PyTorch and TensorFlow by TransformaTech Institute, independently published Nov. 10, 2024.
Other Sources	2. NVIDIA Deep Learning Institute: https://www.nvidia.com/en-us/training/
	3. Deep Learning with Python, Second Edition by Francois Chollet, Publisher: Manning, Dec. 21, 2021.
	4. Deep Learning by Ian Goodfellow, Publisher: The MIT Press, Nov. 18, 2016.
	5. Neural Networks and Deep Learning: A Textbook by Charu C. Aggarwal, Publisher: Springer, Sep. 13, 2018.
	6. PyTorch web page: https://pytorch.org/ & TensorFlow web page: https://www.tensorflow.org/

Evaluation System

Requirements	Number	Percentage of Grade
Attendance/Participation	-	-
Laboratory	-	-
Application	-	-
Field Work	-	-
Special Course Internship	-	-
Quizzes/Studio Critics	-	-
Homework Assignments	1	20
Presentation	-	-
Project	-	-
Report	-	-
Seminar	-	-
Midterms Exams/Midterms Jury	1	35
Final Exam/Final Jury	1	45
Toplam	3	100

Percentage of Semester Work	55
Percentage of Final Work	45
Total	100

Course Category

Core Courses
Major Area Courses	X
Supportive Courses
Media and Managment Skills Courses
Transferable Skill Courses

The Relation Between Course Learning Competencies and Program Qualifications

#	Program Qualifications / Competencies	Level of Contribution
#	Program Qualifications / Competencies	1	2	3	4	5
1	Adequate knowledge in mathematics, science and subjects specific to the computer engineering discipline; the ability to apply theoretical and practical knowledge of these areas to complex engineering problems.					X
2	The ability to identify, define, formulate and solve complex engineering problems; selecting and applying proper analysis and modeling techniques for this purpose.					X
3	The ability to design a complex system, process, device or product under realistic constraints and conditions to meet specific requirements; the ability to apply modern design methods for this purpose.	X
4	The ability to develop, select and utilize modern techniques and tools essential for the analysis and determination of complex problems in computer engineering applications; the ability to utilize information technologies effectively.			X
5	The ability to design experiments, conduct experiments, gather data, analyze and interpret results for the investigation of complex engineering problems or research topics specific to the computer engineering discipline.				X
6	The ability to work effectively in inter/inner disciplinary teams; ability to work individually
7	Effective oral and writen communication skills in Turkish; the ability to write effective reports and comprehend written reports, to prepare design and production reports, to make effective presentations, to give and to receive clear and understandable instructions.
8	The knowledge of at least one foreign language; the ability to write effective reports and comprehend written reports, to prepare design and production reports, to make effective presentations, to give and to receive clear and understandable instructions.
9	Recognition of the need for lifelong learning; the ability to access information, to follow recent developments in science and technology.					X
10	The ability to behave according to ethical principles, awareness of professional and ethical responsibility;
11	Knowledge of the standards utilized in software engineering applications	X
12	Knowledge on business practices such as project management, risk management and change management;
13	Awareness about entrepreneurship, innovation
14	Knowledge on sustainable development
15	Knowledge on the effects of computer engineering applications on the universal and social dimensions of health, environment and safety;
16	Awareness of the legal consequences of engineering solutions
17	An ability to describe, analyze and design digital computing and representation systems.	X
18	An ability to use appropriate computer engineering concepts and programming languages in solving computing problems.					X

ECTS/Workload Table

Activities	Number	Duration (Hours)	Total Workload
Course Hours (Including Exam Week: 16 x Total Hours)	16	3	48
Laboratory
Application
Special Course Internship
Field Work
Study Hours Out of Class	16	2	32
Presentation/Seminar Prepration
Project
Report
Homework Assignments	1	18	18
Quizzes/Studio Critics
Prepration of Midterm Exams/Midterm Jury	1	12	12
Prepration of Final Exams/Final Jury	1	15	15
Total Workload			125