MATH+ML+X Lab Aviles-Rivero Lab

Teaching


[Spring Semester 2025] Machine Learning Theory Course at YMSC, Tsinghua University. https://math-ml-x.github.io/MLT25/ MLT Spring 25 AI Aviles-Rivero
课程号(Course Number): 84760144
课程名(Course Name): 机器学习理论(Machine Learning Theory).
Lecturer: Angelica Aviles-Rivero, YMSC, Tsinghua University
What You’ll Learn
We cover foundational topics like supervised and unsupervised learning, decision theory, and the PAC framework for evaluating model complexity and generalisation. Core concepts such as Empirical Risk Minimisation, regularisation techniques, and the bias-variance trade-off are explored to address challenges like overfitting. Advanced tools, including Rademacher complexities and PAC-Bayesian bounds, provide insights into model generalisation.
Optimisation techniques, from gradient descent to adaptive methods, are examined alongside applications in neural networks and kernel methods. Topics like ensemble learning, probabilistic reasoning, structured prediction, and causal inference round out the curriculum, equipping learners to tackle high-dimensional, complex problems.
Why Choose This Course?
Our course stands out by offering a comprehensive and structured journey into the fascinating world of machine learning. With a focus on both theoretical foundations and practical applications, you’ll gain a profound understanding of why machine learning works and how to build systems that excel. Through a mix of carefully curated readings, interactive discussions, and real-world exercises, we ensure a learning experience that is both engaging and effective.
This course doesn’t just teach you algorithms—it empowers you to think critically about the challenges of data, model generalisation, and optimisation. You’ll master essential concepts such as the PAC framework, VC-dimension, and regularisation, while also exploring advanced topics like neural networks, kernel methods, and probabilistic reasoning. 
Whether you are aiming to strength your knowledge, prepare for research, or innovate in the ML-driven world, this course equips you with the expertise to make a significant impact.Join us this coming Spring 2025!!