Official government website of the Government of Kingdom of Saudi Arabia
Link to official Saudi websites end with edu.sa

All links to official educational websites in the Kingdom of Saudi Arabia end with sch.sa or edu.sa

Government websites use the HTTPS protocol for encryption and security.

Secure websites in the Kingdom of Saudi Arabia use the HTTPS protocol for encryption.

Registered with the Digital Government Authority under number:

20250417892
Main
UQU Programs
Data Science

Optimization for Data Analysis

DS4502
3 hours English

Optimization for Data Analysis

Optimization for Data Analysis
DS3402 Data Modelling
3 hours Two requisites
1 Introduction to Optimization for Data Analysis / Role of optimization in data analysis and machine learning / Overview of optimization problems: convex, non-convex, and stochastic / Real-world application examples
2 Fundamentals of Convex Optimization / Convex sets, functions, and problems / Properties of convex functions and their relevance in data science / Practical examples of convex optimization in data analysis
3 First-Order Optimization Methods / Gradient descent: concepts, implementation, and convergence / Stochastic gradient descent (SGD) for large-scale problems / Proximal algorithms and coordinate descent methods / Applications
4 Second-Order Optimization Methods / Newton's method and quasi-Newton methods / Trust-region and conjugate gradient methods / Trade-offs between first-order and second-order methods / Applications
5 Non-Convex Optimization / Challenges and examples of non-convex optimization / Saddle points, local minima, and global minima / Methods for addressing non-convexity: alternating minimisation and convex relaxations / Applications
6 Stochastic and Online Optimization / Online gradient descent and incremental methods / Scalability for large datasets / Case studies: streaming data and adaptive models
7 Duality and Constrained Optimization / Lagrange duality: theory and applications / KKT conditions and their use in solving constrained problems / Practical examples
8 Min-Max Optimization and Saddle Point Problems / Introduction to min-max problems and saddle-point methods / Extragradient methods and applications / Examples in adversarial machine learning and GANs
9 Distributed and Parallel Optimization / Parallel and distributed methods for large-scale optimization / Synchronous vs. asynchronous communication models / Practical examples
10 Advanced Applications and Case Studies / Optimization in deep learning: backpropagation and optimisation in training / Real-world data analysis tasks: classification, clustering, and recommendation systems / Hands-on lab implementations with Python, R, MATLAB, or other tools
11 Ethical and Practical Considerations / Ethical implications of optimization in data analysis / Fairness in resource allocation and algorithmic decision-making / Communicating optimization results to stakeholders
12 Review, quizzes and midterm
1.1 Mapped to: K2

Explain fundamental optimization concepts, including convex, non-convex, and stochastic optimization, and their relevance to data science applications.

Teaching Strategy Class lectures Active learning Self-learning
Assessment Methods Assignments Exams and quizzes
1.2 Mapped to: K3

Describe the characteristics, applicability, and limitations of first-order and second-order optimization methods.

Teaching Strategy Class lectures Active learning Self-learning
Assessment Methods Assignments Exams and quizzes
1.3 Mapped to: K2

Understand the role of optimization techniques in solving data-intensive problems such as model training, feature selection, and resource allocation.

Teaching Strategy Class lectures Active learning Self-learning
Assessment Methods Assignments Exams and quizzes
2.1 Mapped to: S1

Formulate real-world problems as mathematical optimization tasks and select appropriate solution methods based on the problem type.

Teaching Strategy Class lectures Active learning Self-learning
Assessment Methods Assignments Exams and quizzes Project
2.2 Mapped to: S2

Implement optimization algorithms using Python and R libraries, such as SciPy, and adapt them to data-driven challenges.

Teaching Strategy Class lectures Active learning Self-learning Projects-based learning Cooperative learning Labs
Assessment Methods Assignments Exams and quizzes Project
2.3 Mapped to: S3

Analyse the scalability and efficiency of optimization methods when applied to large-scale datasets or complex problems.

Teaching Strategy Class lectures Active learning Self-learning Projects-based learning Cooperative learning Labs
Assessment Methods Assignments Exams and quizzes Project
2.4 Mapped to: S2

Solve practical optimization problems in data science workflows, including machine learning and operational decision-making.

Teaching Strategy Class lectures Active learning Self-learning Projects-based learning Cooperative learning Labs
Assessment Methods Assignments Exams and quizzes Project
3.1 Mapped to: V2

Demonstrate independent problem-solving skills by applying optimization techniques to real-world scenarios.

Teaching Strategy Class lectures Active learning Self-learning Projects-based learning
Assessment Methods Assignments Project
3.2 Mapped to: V2

Exhibit accountability in using optimization methods ethically and responsibly in contexts involving data analysis and resource allocation.

Teaching Strategy Class lectures Active learning Self-learning Projects-based learning
Assessment Methods Assignments Project
3.3 Mapped to: V3

Collaborate effectively within teams to address optimization challenges, valuing diverse approaches and perspectives.

Teaching Strategy Class lectures Active learning Self-learning Projects-based learning
Assessment Methods Assignments Project
Assessment
Weight
Assignments
10%
Group project
40%
Quizzes
10%
Final exam
40%
Total
100%