Spring 2025:
As machine learning increasingly pervades more and more aspects of our life, hitherto often ignored questions regarding social responsibility, transparency, and trustworthiness of machine learning and AI emerge as mainstream topics that pose deep and challenging mathematical problems. How do we take fairness and transparency into account while developing machine-learned models and systems? How do we protect the privacy of users when building large-scale, AI-based systems? How can we develop a rigorous understanding of the vulnerabilities inherent to machine learning? Due to their real-world implications, these topics are now at the forefront of ML and AI research. This course will introduce and analyze the mathematical concepts behind fairness and privacy of machine learning. Benefits and shortcomings of existing mathematical tools, metrics, and methods will be investigated and open problems will be discussed.
Prerequisites: Linear algebra and a basic background in probability as well as basic experience in programming will be required. Some basic knowledge in optimization and machine learning is helpful.
For more information see the Canvas course page.
Introduction to the mathematics of time-frequency analysis and multiscale analysis (wavelets, Gabor frames, and their relatives) useful for diverse fields such as signal and image processing, machine learning, quantum physics, numerical analysis, and statistics.
For more information see the Canvas course page.