Applied Machine Learning in Python

Course Description

Real-world applications of machine learning require not only a strong theoretical foundation but also a solid knowledge of the methodologies, tools, and heuristics essential for implementing machine learning algorithms. However, the practical aspects of machine learning are often overlooked in mathematics programs. This course bridges that gap by providing students with hands-on experience in implementation and empirical analysis of machine learning algorithms — critical skills for those pursuing careers in data analysis or machine learning.

Content

The course covers fundamental topics such as linear regression, gradient descent, regularization techniques, logistic regression, support vector machines (SVMs), and basic neural networks. Additionally, the course will explore advanced optimization methods, multi-class classification strategies, and ensemble learning techniques such as boosting and bagging.

A key component of the course is extensive programming in Python, using key libraries such as NumPy, Matplotlib, Pandas, and scikit-learn. We will work with real datasets, including MNIST handwritten digits, the Boston Housing dataset, Wine dataset, etc.

The materials are publicly available on the course's GitHub repository: https://github.com/mselezniova/AppliedML

Schedule and Venue

• Lecture (by Mariia Seleznova, Dr.): Tue 10-12 in room B 046.
• Exercise (by Mariia Seleznova, Dr.): Wed 10-12 in room B 046.

Creditable Modules

• MSc FiMa: WP23 „Advanced Topics in Computer and Data Science B”
• MSc Math: WP42 „Überblick über ein aktuelles Forschungsgebiet B”

Other modules and MSc programs are possible as well. Interested students should directly approach the Prüfungsamt and inform us.

Prerequisites

The course is targeted at mathematics MSc students, good knowledge of linear algebra, probability, and statistics is required. Basic knowledge of machine learning (statistical learning) theory and optimization is recommended.

Registration

Please subscribe to the course Moodle page here: https://moodle.lmu.de/course/view.php?id=45008 (password: aml26).

The number of participants is limited to 30, as the course is graded through projects completed in pairs. Since there were more applicants than available places last year, it is possible that I will not be able to admit everyone who registers on Moodle. To reserve a spot, you may email me in advance at selez@math.lmu.de (only if you plan to participate actively in the course!). Any remaining places will be allocated during the first lecture on April 14, on a first-come, first-served basis.