Professional Master's in Artificial Intelligence & Data Sciences

A professional master's program at the Faculty of Artificial Intelligence Engineering: two academic years over four semesters, from AI and data engineering foundations to generative and agentic AI, ending in a research graduation project.

Degree
Professional Master's
Duration
2 years (4 semesters)
Credit hours
100 minimum
Courses
23 courses

The Faculty of Artificial Intelligence Engineering offers a Professional Master's Degree Program in Artificial Intelligence & Data Sciences.

On the proposal of the Faculty Council, Syrian Private University awards the Professional Master's Degree in Artificial Intelligence & Data Sciences, in addition to the bachelor's degrees the faculty awards.

Study lasts two academic years over four levels (semesters), with a minimum of 100 credit hours.

Study plan

Select a course to see its description, learning outcomes and prerequisites.

Total: 100 credit hours · 23 courses

Year 1 · Semester 1

Foundations

Total: 24 credit hours · Theory 18 · Practical 12

Year 1 · Semester 2

Core AI & Data Engineering

Total: 24 credit hours · Theory 18 · Practical 12

Year 2 · Semester 3

Advanced AI & Specializations

Total: 24 credit hours · Theory 18 · Practical 12

Year 2 · Semester 4

Research & Graduation Project

Total: 28 credit hours · Theory 19 · Practical 18

Course codes: two letters for the program (MADS), one digit for the level, and two digits for the course sequence (01 to 99).

MADS101

Advanced Python Programming for AI

البرمجة المتقدمة بلغة بايثون للذكاء الاصطناعي

credit hours
4
Theory
3
Practical
2

Covers Python 3 advanced features, OOP, functional programming, asynchronous programming, NumPy, Pandas, Matplotlib, and best practices for scientific computing. Emphasizes writing efficient, testable, and production-ready AI code.

Prerequisites

None

Opens

Course learning outcomes

  1. Write and debug advanced Python scripts using OOP and functional paradigms.
  2. Manipulate and visualize large datasets using NumPy, Pandas, and Matplotlib.
  3. Develop asynchronous and parallel Python applications for data pipelines.
  4. Apply software engineering best practices to AI project development.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS102

Advanced Algorithms & Data Structures for AI

الخوارزميات وبنى المعطيات المتقدمة للذكاء الاصطناعي

credit hours
4
Theory
3
Practical
2

Advanced algorithm design paradigms with a focus on AI-specific algorithms: A* and heuristic search, minimax and alpha-beta pruning, Monte Carlo Tree Search, constraint satisfaction, genetic algorithms, simulated annealing, ant colony and swarm intelligence, approximation algorithms, and NP-hard problem handling. Data structures critical for AI: priority queues, tries, graphs, KD-trees, and spatial indexing.

Prerequisites

None

Opens

Course learning outcomes

  1. Implement and compare heuristic search algorithms including A* and its variants.
  2. Apply adversarial search and game-tree algorithms to AI decision problems.
  3. Design and evaluate metaheuristic and evolutionary algorithms for optimization.
  4. Select and implement advanced AI-relevant data structures for efficient computation.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS103

Mathematical Foundations for AI

الأسس الرياضية للذكاء الاصطناعي

credit hours
4
Theory
3
Practical
2

A unified course integrating probability theory, statistics, and numerical methods essential for AI. Topics include: random variables, distributions, Bayesian inference, hypothesis testing, regression analysis, multivariate statistics, linear algebra for ML, numerical methods for solving ODEs/PDEs, root-finding, interpolation, numerical integration, gradient-based and convex optimization, and metaheuristic optimization techniques.

Prerequisites

None

Opens

Course learning outcomes

  1. Apply probability distributions and Bayesian inference to model AI problems.
  2. Perform statistical analysis and hypothesis testing relevant to data science.
  3. Apply linear algebra and numerical methods to AI computations.
  4. Implement and compare gradient-based and metaheuristic optimization algorithms.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS104

Advanced Databases for AI

قواعد البيانات المتقدمة للذكاء الاصطناعي

credit hours
4
Theory
3
Practical
2

Relational database design, SQL optimization, data warehousing, OLAP, introduction to NewSQL systems, and database architectures for AI workloads including in-memory databases and time-series databases.

Prerequisites

None

Opens

Course learning outcomes

  1. Design and optimize relational schemas for analytical workloads.
  2. Implement data warehousing and OLAP solutions.
  3. Compare and select database technologies for various AI use cases.
  4. Apply SQL and query optimization techniques for large datasets.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS105

Distributed Systems & Cloud Computing

الأنظمة الموزعة والحوسبة السحابية

credit hours
4
Theory
3
Practical
2

Principles of distributed computing, consistency models, fault tolerance, microservices, containerization (Docker, Kubernetes), cloud platforms (AWS, GCP, Azure), serverless computing, and cloud-native AI deployment.

Prerequisites

None

Course learning outcomes

  1. Explain distributed systems concepts including consistency and fault tolerance.
  2. Deploy containerized AI applications using Docker and Kubernetes.
  3. Design cloud-native architectures for AI workloads on major platforms.
  4. Implement scalable data pipelines using cloud services.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS106

Natural Language Processing

معالجة اللغات الطبيعية

credit hours
4
Theory
3
Practical
2

Text preprocessing, tokenization, word embeddings (Word2Vec, GloVe, FastText), sequence models, BERT, GPT architectures, named entity recognition, sentiment analysis, machine translation, and text summarization.

Prerequisites

None

Opens

Course learning outcomes

  1. Implement text preprocessing and feature extraction pipelines.
  2. Apply pre-trained language models for downstream NLP tasks.
  3. Fine-tune transformer models for specific NLP applications.
  4. Evaluate NLP systems using appropriate metrics.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS201

Machine Learning

تعلم الآلة

credit hours
4
Theory
3
Practical
2

Supervised and unsupervised learning, regression, classification, clustering, dimensionality reduction, ensemble methods, model evaluation, cross-validation, bias-variance tradeoff, and feature engineering.

Prerequisites

Course learning outcomes

  1. Implement and evaluate supervised learning algorithms.
  2. Apply unsupervised techniques for clustering and dimensionality reduction.
  3. Design feature engineering pipelines for diverse data types.
  4. Evaluate and tune ML models using rigorous validation techniques.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS202

Deep Learning

التعلم العميق

credit hours
4
Theory
3
Practical
2

Neural network fundamentals, backpropagation, CNNs, RNNs, LSTMs, Transformers, attention mechanisms, transfer learning, regularization, advanced optimization, and deployment with TensorFlow and PyTorch.

Prerequisites

Opens

Course learning outcomes

  1. Build and train CNNs and RNNs for vision and sequence tasks.
  2. Implement Transformer architectures from scratch.
  3. Apply transfer learning and fine-tuning strategies.
  4. Deploy deep learning models using TensorFlow and PyTorch.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS203

NoSQL & Big Data Technologies

تقنيات NoSQL والبيانات الكبيرة

credit hours
4
Theory
3
Practical
2

NoSQL paradigms (document, key-value, column-family, graph), MongoDB, Cassandra, Redis, Neo4j, Apache Hadoop, HDFS, MapReduce, Apache Spark, Kafka, and data lake architectures for AI workloads.

Prerequisites

Opens

Course learning outcomes

  1. Design and implement NoSQL database solutions for AI applications.
  2. Build and operate Apache Spark pipelines for large-scale data processing.
  3. Design streaming data architectures using Kafka.
  4. Implement data lake solutions and govern data quality.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS204

Image Processing & Computer Vision

معالجة الصور والرؤية الحاسوبية

credit hours
4
Theory
3
Practical
2

Image fundamentals, filtering, morphological operations, feature extraction (SIFT, HOG), object detection (YOLO, SSD, Faster R-CNN), image segmentation, pose estimation, 3D vision, and OpenCV with deep learning frameworks.

Prerequisites

Opens

Course learning outcomes

  1. Apply classical image processing techniques for feature extraction.
  2. Implement object detection and segmentation models.
  3. Develop end-to-end computer vision pipelines.
  4. Evaluate computer vision systems on benchmark datasets.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS205

AI High Performance Computing

الحوسبة عالية الأداء للذكاء الاصطناعي

credit hours
4
Theory
3
Practical
2

GPU architecture, CUDA programming, parallel computing models, distributed training (data parallelism, model parallelism), mixed-precision training, model compression, ONNX, TensorRT, and profiling AI workloads.

Prerequisites

Course learning outcomes

  1. Program GPU kernels for parallel AI computations using CUDA.
  2. Implement distributed training strategies for large models.
  3. Optimize AI models using quantization and pruning techniques.
  4. Profile and benchmark AI system performance.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS206

Graph Machine Learning

تعلم الآلة على البيان

credit hours
4
Theory
3
Practical
2

Graph theory for ML, node embeddings (DeepWalk, Node2Vec), graph neural networks (GCN, GAT, GraphSAGE), knowledge graphs, link prediction, community detection, and applications in drug discovery, social networks, and recommendation systems.

Prerequisites

Opens

Course learning outcomes

  1. Implement graph neural network architectures for node and graph tasks.
  2. Apply graph embeddings for representation learning.
  3. Build knowledge graph applications.
  4. Deploy GML solutions for real-world domain problems.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS301

Generative AI

الذكاء الاصطناعي التوليدي

credit hours
4
Theory
3
Practical
2

Variational Autoencoders, GANs, diffusion models, large language models (LLMs), GPT-4/Claude/Gemini architectures, prompt engineering, RLHF, fine-tuning LLMs, multimodal generation, and AI content safety.

Prerequisites

Opens

Course learning outcomes

  1. Implement and compare GAN, VAE, and diffusion model architectures.
  2. Apply prompt engineering and fine-tuning techniques to LLMs.
  3. Develop multimodal generative AI applications.
  4. Evaluate generative models for quality, safety, and fairness.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS302

Agentic AI & Multi-Agent Systems

الذكاء الاصطناعي الوكيل وأنظمة متعددة الوكلاء

credit hours
4
Theory
3
Practical
2

AI agent architectures (ReAct, Plan-and-Execute), tool use, memory systems, LangChain, LlamaIndex, AutoGen, CrewAI, multi-agent coordination, planning under uncertainty, and autonomous AI systems deployment.

Prerequisites

Opens

Course learning outcomes

  1. Design and implement AI agent pipelines using LangChain or equivalent.
  2. Build multi-agent systems for complex task decomposition.
  3. Integrate external tools and APIs into agentic workflows.
  4. Evaluate reliability, safety, and performance of autonomous AI systems.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS303

Advanced Deep Learning

التعلم العميق المتقدم

credit hours
4
Theory
3
Practical
2

Advanced deep learning topics beyond fundamentals: Parameter-Efficient Fine-Tuning (PEFT) — LoRA, QLoRA, Prefix Tuning, Adapters; self-supervised and contrastive learning (SimCLR, MoCo, BYOL); neural architecture search (NAS); mixture of experts (MoE); state space models (Mamba); graph neural networks for DL; meta-learning and few-shot learning; neural ODEs; and scalable training strategies for very large models.

Prerequisites

Opens

Course learning outcomes

  1. Implement PEFT techniques (LoRA, QLoRA, Adapters) for efficient LLM fine-tuning.
  2. Apply self-supervised and contrastive learning frameworks.
  3. Explore advanced architectures including MoE, state space models, and meta-learning.
  4. Design scalable training strategies for very large deep learning models.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS304

TinyML & Edge AI

الذكاء الاصطناعي على الأجهزة الطرفية (TinyML)

credit hours
4
Theory
3
Practical
2

Model quantization, pruning, knowledge distillation, neural architecture search, TensorFlow Lite, ONNX Runtime, deployment on microcontrollers (Arduino, STM32, Raspberry Pi), Edge Impulse, and federated learning.

Prerequisites

Opens

Course learning outcomes

  1. Apply model compression techniques for edge deployment.
  2. Deploy AI models on microcontrollers and embedded systems.
  3. Implement federated learning for privacy-preserving AI.
  4. Benchmark TinyML models for accuracy-latency tradeoffs.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS305

Vector Databases & Retrieval-Augmented Generation

قواعد البيانات المتجهة والتوليد المعزز بالاسترجاع

credit hours
4
Theory
3
Practical
2

Vector embeddings, similarity search, ANN algorithms (HNSW, IVF), vector databases (Pinecone, Weaviate, Milvus, Chroma), RAG architectures, hybrid search, chunking strategies, and production RAG system design.

Prerequisites

Opens

Course learning outcomes

  1. Design and implement vector indexing and similarity search systems.
  2. Build end-to-end RAG pipelines with LLMs and vector stores.
  3. Evaluate and optimize retrieval quality in RAG systems.
  4. Deploy production-grade vector database solutions.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS306

Multimodal AI & Cross-Modal Learning

الذكاء الاصطناعي متعدد الوسائط والتعلم عبر الأنماط

credit hours
4
Theory
3
Practical
2

Foundations of multimodal learning; vision-language models (CLIP, ALIGN, Florence); image captioning and visual question answering (VQA); text-to-image generation (Stable Diffusion, DALL-E); audio-visual models; multimodal transformers (Flamingo, LLaVA, GPT-4V); cross-modal retrieval; multimodal fusion strategies; video understanding; and evaluation metrics for multimodal systems.

Prerequisites

Opens

Course learning outcomes

  1. Implement vision-language models such as CLIP for cross-modal tasks.
  2. Build multimodal pipelines for VQA, image captioning, and text-to-image generation.
  3. Apply multimodal fusion strategies and cross-modal retrieval techniques.
  4. Evaluate multimodal AI systems using domain-appropriate metrics.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS401

AI Ethics, Governance & Research Methodology

أخلاقيات الذكاء الاصطناعي والحوكمة ومنهجية البحث

credit hours
4
Theory
3
Practical
2

Research design, literature review, scientific writing, AI ethics frameworks (fairness, accountability, transparency), algorithmic bias, data privacy (GDPR, AI Act), AI governance policies, and responsible deployment practices.

Prerequisites

Course learning outcomes

  1. Conduct systematic literature reviews in AI and data science.
  2. Apply ethical frameworks to evaluate AI system designs.
  3. Communicate research findings through scientific writing.
  4. Analyze AI governance and compliance requirements.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS402

Advanced Topics in AI (Seminar)

موضوعات متقدمة في الذكاء الاصطناعي (سيمينار)

credit hours
8
Theory
6
Practical
4

Current research frontiers in AI including foundation models, world models, neuromorphic computing, AI for science (drug discovery, climate, genomics), AI safety and alignment, and emerging paradigms. Includes invited lectures and student presentations.

Prerequisites

Course learning outcomes

  1. Critically review and present recent AI research papers.
  2. Identify open research problems in AI and data science.
  3. Synthesize knowledge across AI sub-fields for cross-domain applications.
  4. Communicate advanced AI concepts to diverse audiences.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS403

AI Product Engineering

هندسة منتجات الذكاء الاصطناعي

credit hours
4
Theory
3
Practical
2

End-to-end AI product lifecycle: experiment tracking (MLflow, W&B), data versioning (DVC), CI/CD for ML, feature stores, model serving (FastAPI, BentoML, Triton), A/B testing, shadow and canary deployments, drift detection, model monitoring, AI security (adversarial attacks, model stealing, prompt injection), compliance engineering, and industry case studies in production AI.

Prerequisites

Course learning outcomes

  1. Design and implement full MLOps pipelines with experiment tracking and CI/CD.
  2. Deploy and monitor AI models at scale with drift detection and alerting.
  3. Implement AI security measures against adversarial and injection threats.
  4. Apply responsible AI practices in production product development.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS404

Reinforcement Learning & Decision Making

التعلم المعزز واتخاذ القرار

credit hours
4
Theory
3
Practical
2

Markov Decision Processes, dynamic programming, Monte Carlo methods, temporal difference learning, Q-learning, Deep Q-Networks (DQN), policy gradient methods, Proximal Policy Optimization (PPO), multi-armed bandits, and applications in robotics and game playing.

Prerequisites

Course learning outcomes

  1. Model sequential decision problems as MDPs.
  2. Implement value-based and policy-based RL algorithms.
  3. Apply deep RL to control and game-playing environments.
  4. Evaluate RL agent performance and convergence.

Grading

Final exam 50%, projects/assignments 35%, quizzes and participation 15%.

MADS405

Master Graduation Project

مشروع تخرج الماجستير

credit hours
8
Theory
4
Practical
8

Literature review, problem formulation, dataset collection, methodology design, preliminary experiments, full implementation, experimentation, results analysis, and academic paper writing. Supervised by a faculty advisor.

Prerequisites

All Semester 1-3 courses

Course learning outcomes

  1. Define a clear AI/data science research problem and objectives.
  2. Conduct a comprehensive literature review.
  3. Design an appropriate research methodology, implement and validate the proposed AI/data science solution, and analyze and critically discuss experimental results.
  4. Write a complete Master's thesis adhering to academic standards and defend it professionally before an examination committee.

Grading

Final presentation 40%, thesis 30%, Seminar 2 15%, Seminar 1 15%.

Program learning outcomes

  1. PLO1 Apply advanced mathematical, statistical, and computational foundations to solve complex AI and data science problems.
  2. PLO2 Design and implement machine learning and deep learning models for real-world applications.
  3. PLO3 Develop and deploy generative AI and agentic AI systems using state-of-the-art architectures.
  4. PLO4 Engineer scalable big data pipelines and manage diverse data storage paradigms including NoSQL and Vector databases.
  5. PLO5 Apply natural language processing and computer vision techniques to extract insights from unstructured data.
  6. PLO6 Optimize AI workloads on high-performance computing and edge/embedded environments.
  7. PLO7 Evaluate, select, and integrate appropriate AI and data science tools, frameworks, and platforms for diverse engineering domains.
  8. PLO8 Conduct original research and produce a professional Master's thesis demonstrating mastery of AI and data science methodologies.
  9. PLO9 Communicate technical findings effectively to both technical and non-technical stakeholders in professional settings.
  10. PLO10 Apply ethical, legal, and societal considerations in the development and deployment of AI systems.

Curriculum coverage matrix

I = Introduced, R = Reinforced, M = Mastered

Course PLO1 PLO2 PLO3 PLO4 PLO5 PLO6 PLO7 PLO8 PLO9 PLO10
MADS101 Advanced Python Programming for AI I I I
MADS102 Advanced Algorithms & Data Structures for AI I I I
MADS103 Mathematical Foundations for AI I I I
MADS104 Advanced Databases for AI I I I
MADS105 Distributed Systems & Cloud Computing I I I
MADS106 Natural Language Processing I I I
MADS201 Machine Learning R I R
MADS202 Deep Learning R R R
MADS203 NoSQL & Big Data Technologies R R
MADS204 Image Processing & Computer Vision R R R
MADS205 AI High Performance Computing R R
MADS206 Graph Machine Learning M R M
MADS301 Generative AI M I R
MADS302 Agentic AI & Multi-Agent Systems R M R
MADS303 Advanced Deep Learning M R R
MADS304 TinyML & Edge AI M M
MADS305 Vector Databases & Retrieval-Augmented Generation R M R
MADS306 Multimodal AI & Cross-Modal Learning R M M
MADS401 AI Ethics, Governance & Research Methodology I R M
MADS402 Advanced Topics in AI (Seminar) R M R
MADS403 AI Product Engineering M M M M M
MADS404 Reinforcement Learning & Decision Making M R R
MADS405 Master Graduation Project M M M M R

Assessment strategies

Final Examination

50%

Comprehensive written exam covering all course material.

Projects & Practical Work

35%

Individual and group projects, lab assignments, Jupyter notebooks, GitHub submissions, presentations.

Quizzes, Homework & Participation

15%

Regular quizzes, homework assignments, class participation, and peer reviews.

Graduation Project Defense

Semester 4

Oral defense before a three-member examination committee. Graded on originality, methodology, results, and presentation quality.

Master's graduation project rules

The Master's graduation project is a course like any other. It teaches students to tackle a specialized engineering topic in artificial intelligence and data science and to implement the solution they reach. The project is carried out as follows:

  1. 01 The student prepares a project on a topic approved by the Faculty Council and submits a report on the work, which is defended before an examination committee the Faculty Council forms for this purpose.
  2. 02 The graduation project must fall within the program's specializations.
  3. 03 The project is supervised by a member of the teaching staff or equivalent, or by a PhD-holding specialist approved by the Faculty Council.
  4. 04 To register for the Master's project, the student must have completed at least 72 credit hours.
  5. 05 Master's graduation projects are assessed by a three-member committee formed by the Faculty Council, on dates the Council sets.
  6. 06 The committee grades the project under rules set by the Faculty Council. The student passes with a mark of at least 60 out of 100.
  7. 07 When necessary, the committee may allow up to one month from the defense date for revisions; the defense is not repeated in that case.
  8. 08 A student fails the project with a mark below 60, or if the committee does not accept the requested revisions within one month of the defense. The student may then register for the project again, once.

Admission requirements

Candidates for the Professional Master's in Artificial Intelligence & Data Sciences must hold a bachelor's degree with at least a Good graduation average from a public or private university, the technology programs of the Syrian Virtual University, an applied college, or a recognized non-Syrian university, in one of the following majors:

Eligible majors

  • Informatics Engineering
  • Mechanical and Electrical Engineering (Computer Engineering)
  • Computer and Informatics Engineering
  • Information and Communication Technology Engineering (Information Technology)
  • Informatics and Communications Engineering (Information Systems, Software, Artificial Intelligence)
  • Software Engineering and Information Systems
  • Electrical Engineering (Automatic Control and Computer Engineering)
  • Electronic Engineering (Computer Engineering, Computers and Automatic Control)
  • Science (Informatics, Applied Mathematics and Informatics)
  • Information Engineering (Information Technology)
  • Engineering Diploma of the Higher Institute for Applied Sciences and Technology (Informatics / Information Systems Engineering)
  • Computer Science
  • Computer Systems Engineering
  • Information Systems / Information Systems and Technology Management / Management Information Systems (MIS)
  • Bachelor's in Informatics (UOG)
  • Arab Academy for Science, Technology and Maritime Transport (Computer Engineering)
  • Computer Engineering (Automation and Control) / Communications Engineering / Information and Communications Engineering / Network Engineering / Biomedical Engineering

Ranking criteria

Graduation average 80%

The graduation average of the degree on which the candidate is admitted.

English language test 20%

The result of the English language test held by the university.

The University Council may adjust admission averages at the start of each semester, according to the approved intake policy and the university's capacity.

Degree requirements

  1. Pass all courses.
  2. Pass the Master's project.
  3. Pay the fees and installments due under the university's financial regulations.

Teaching and assessment

Teaching, examinations and assessment follow Higher Education Council Decision No. 21 of 28/9/2014 and its amendments, the unified system for Syrian private universities that use credit hours.