Course: Introduction to Learning Systems and Data Science
See here for the current iteration of this course.
This course, part of the curriculum for aspiring data scientists and AI practitioners, offers a foundational understanding of key concepts and methodologies in learning systems and data science. The primary focus is on learning systems, their architectures, and how they can be effectively designed and implemented to solve complex problems.
Students will begin with Linear Algebra and Statistics for Data Science, which are crucial for understanding data structures and algorithms used in learning systems. This mathematical foundation will support later explorations into model evaluation and performance assessment.
The course will comprehensively cover Optimization Techniques for Learning, where students will learn how to enhance learning systems through parameter tuning and various optimization algorithms. Understanding how to optimize models is essential for building high-performing learning systems that can adapt over time.
In addition to foundational concepts, participants will gain insights into the Fundamentals of Machine Learning, focusing on how different algorithms can be applied to create learning systems that make predictions and automate decision-making processes. The curriculum will also explore Neural Networks and Deep Learning, emphasizing their pivotal role in the development of advanced learning systems capable of handling intricate tasks, such as image and speech recognition.
The course will introduce Natural Language Processing and Language Models, which will allow students to examine how learning systems can understand and generate human language. This segment will also cover practical applications, such as chatbots and sentiment analysis tools.
An important aspect of the course will be Hybrid Learning Approaches, highlighting how different machine learning paradigms (e.g., supervised, unsupervised, and reinforcement learning) can be combined to build more robust and flexible learning systems.
Human-centred Design principles will be emphasized to ensure that learning systems are intuitive, user-friendly, and effectively meet user needs. The course will also delve into the ethical implications of developing and deploying learning systems, covering considerations such as bias, fairness, and transparency in AI.
Practical Examples will serve as a capstone to the course, where students will engage in hands-on projects that reinforce learning. These projects will allow participants to apply the theoretical concepts learned throughout the course to build real-world learning systems that can analyze data, make decisions, and interact with users effectively.
Course Topics
- Linear Algebra and Statistics for Data Science
- Optimization Techniques for Learning
- Fundamentals of Machine Learning
- Neural Networks and Deep Learning
- Natural Language Processing and Language Models
- Introduction to Hybrid Learning Approaches
- Human-centred Design
- Ethics and Deployment in Learning Systems
- Practical Examples
Course: Mining Media Data I
See here for the current iteration of this course.
This course, offered as part of the Master’s Program in Media Informatics at the Bonn-Aachen International Center for Information Technology (B-IT), provides a comprehensive exploration of advanced data mining techniques tailored for media data analysis. Students will delve into methods like affinity mining, latent pattern mining, neural networks, and archetypal analysis to uncover insights in behavioral profiling, recommender systems, and outlier detection. Emphasis is placed on theoretical understanding and practical application through mathematical optimization, interpretable models, and real-world case studies, enabling participants to harness data for impactful digital marketing, fraud detection, and content personalization.
Course Topics
- Analyze and extract meaningful relationships from large-scale media datasets using advanced data mining techniques.
- Develop and implement predictive and descriptive models for applications such as recommender systems, trend analysis, and outlier detection.
- Apply mathematical optimization methods to create interpretable and efficient machine learning models.
- Integrate theoretical concepts with practical tools to address challenges in digital forensics, behavioral profiling, and marketing strategy design.
Course: Mining Media Data II
This course explores advanced techniques in data mining, emphasizing predictive and prescriptive methods applied to media data. Students will learn to analyze large and complex datasets using state-of-the-art machine learning methodologies, including behavioral prediction, knowledge distillation, and large language models (LLMs). The curriculum includes foundational concepts, text representation learning, transformer architectures, and practical applications in media analytics, such as recommendation systems and information extraction.
Course Topics
- Understand and implement advanced data mining techniques for predictive and prescriptive analytics.
- Employ large language models and transformer-based architectures for tasks like text analysis, classification, and summarization.
- Apply knowledge distillation techniques to optimize and deploy machine learning models in resource-constrained environments.
- Analyze media data effectively to derive insights and support decision-making in real-world applications, including digital marketing and fraud detection.
- Address challenges in media analytics, such as ethical considerations, model interpretability, and efficient resource use.
Lab: Explainable AI and Applications
In this lab “Explainable AI and Applications – Explainability of foundation models for sequential data”, we will start with the reproduction of existing explainability of deep-learning systems (especially foundation models) in the fields of biomedicine and natural language processing. Then, we will encourage lab participants to find limitations and explore novel solutions with experiments. The students will work in groups on a selected task.
Lab Activities
- Understanding the Landscape: Initiating the course with a comprehensive survey of explainable AI.
- Reproducing Key Findings: Students will select research papers to reproduce significant findings.
- Midterm Milestone: A mid-term presentation to share progress, challenges, and insights.
- Forging New Paths: Groups develop new ideas, benchmarks, datasets, or methodologies.
- Showcasing Innovations: Teams present their original contributions.
- Reflecting and Reporting: The course concludes with a reflection and reporting phase.
Lab: Hybrid Learning and Applications
This lab offers a comprehensive introduction to hybrid learning, merging machine learning and deep learning techniques to address complex problems. Students explore a range of fascinating applications and are encouraged to select and research their project topics.
Lab Activities
- Independent Research and Knowledge Acquisition: Students study a self-selected research topic.
- Practical Application: Apply theoretical knowledge to real-world problems.
- Communication Skills: Structured presentations in oral and written form.
Seminar: Theory of Deep Learning
This seminar focuses on the theoretical underpinnings of deep neural networks, particularly exploring the infinite width limit.
Seminar Topics
- Introduction to Neural Network Theory
- Gaussian Processes and Kernel Learning
- Neural Tangent Kernel and Training Dynamics
- Tensor Programs in Neural Networks
- Generalizability and Theoretical Applications
Seminar: Data Science for Medical Applications
In this seminar, current, relevant research progress in the field of data science in the analysis of medical data is reviewed and presented by the students.
Seminar Topics
- Familiarization with different types of biomedical data.
- Exploration of benefits data science can offer to medicine.
- Overview of latest research and technological advancements.
- Enhancement of critical thinking and problem-solving skills.