Courses

We offer this course from Winter Semester 2023/24 onwards. The lecture content covers the theory of Machine Learning for Business Applications and required fundamentals.

To participate in this course, you have to enroll via TUMonline.

Mandatory Prerequisites

• Mathematics in Natural and Economic Science 1
• Statistics for Business Administration

Intended Learning Outcomes

After participating in this lecture, students have a basic knowledge in the domain of machine learning. Moreover, they have an overview of recent developments and topics. They are able to apply a machine learning framework to a practical problem, know the advantages and disadvantages of various methods and are able to identify and circumvent typical pitfalls.

Course Organization

Topics include but are not limited to:

• Naive Bayes & Bayesian Networks
• Decision Trees
• Ensemble Methods & Clustering
• Regression & Causal Inference
• Data Preparation, Generalization & Evaluation

We offer the course Introduction to Deep Reinforcement Learning from Winter Semester 2021/22 onwards. The course covers rational decision making on the intersection of Operations Research and Machine Learning with Deep Reinforcement Learning. 

To participate in this course, you have to enroll via TUMonline.

Mandatory Prerequisites

To successfully attend this course, students should be comfortable with math-centric content, algorithms, and proofs. Students should have a general understanding of:

• basic linear algebra, including for example matrix multiplication and matrix-vector multiplication
• multivariate calculus, including for example partial derivatives, the chain rule, and gradients
• basic stochastics, including for example discrete and continuous random variables and probability distributions, as well as the notions of expectation and variance
• basics of mathematical optimization, including for example constrained optimization problems and the notion of convergence

For the programming exercises, which are a part of this course and its exam, we use the Python programming language and the NumPy library. Thus, students should ideally be familiar with Python. Alternatively, knowledge of a general purpose programming language (e.g., C++, Java) or Matlab is sufficient as well, as students will be able to adapt to Python very quickly.

Intended Learning Outcomes

After attending this course, students will have acquired:

• basic knowledge in the domain of search algorithms, e.g., graph and tree search, and understand the fundamental theory behind it
• the competence/capability to analyze a practical problem by modelling it as a Markov Decision Process (MDP)
• profound knowledge in the domain of reinforcement learning and understanding of fundamental reinforcement learning theory, e.g., Q-learning, TD learning
• basic knowledge in deep learning and understanding of fundamental machine learning and deep learning theory, e.g., stochastic gradient descent, logistic regression, artificial neural networks
• profound knowledge in the domain of deep reinforcement learning (DRL) that combines the previous two competence areas and understanding of fundamental DRL theory, e.g., deep Q-networks (DQN), advanced policy gradient methods such as proximal policy optimization (PPO)
• the competence/capability to apply a DRL framework to a practical problem
• the competence/capability to evaluate DRL methods w.r.t. to advantages and disadvantages
• the competence/capability to evaluate practical applications w.r.t. typical pitfalls (e.g., convergence issues with non-independent samples) when using DRL and how to circumvent them

Course Organization

The module content covers the theory of Deep Reinforcement Learning and required fundamentals. Specifically, topics include but are not limited to:

• fundamentals (e.g., stochastic gradient descent, logistic regression, artificial neural networks)
• Deep Q-Networks and Rainbow DQNs
• policy gradients, trust region policy optimization, proximal policy optimization
• actor critic methods, soft actor critic methods
• applied case studies

We offer this seminar from Summer Semester 2023 onwards. In this seminar, we build on the theoretical foundations that we learned in the course "Introduction to Deep Reinforcement Learning" and study a real-world application case in order to understand how to design and implement advanced DRL algorithms to solve use cases of practical interest.

To participate in this course, you have to enroll via TUMonline.

Mandatory Prerequisites

• Profound programming skills and at least intermediate knowledge in Python
• Passed exam of Introduction to Deep Reinforcement Learning (MGT001299)

Intended Learning Outcomes

The objective of this seminar is to provide the students with the necessary skills to conduct independent research in the field of Deep Reinforcement Learning, e.g., in preparation for a successful master thesis or a research stay abroad. Herein, students learn how to i) structure a research question and conduct a literature review, ii) use and develop adequate models, software, and algorithms to solve the research question, iii) write a research paper, especially how to structure and organize results and the corresponding methodology, and iv) work in an academic environment where knowledge is mostly gained from recent papers and less from standardized books.

Teaching and Learning Method

This seminar bases on various methods: First, the lecturer gives an introduction into the general topic and an overview on recent trends and challenges in an introductory session. Afterward, a case study based on a real-world application is introduced. Then, students will work in small groups on this case study, using and practicing their skills in literature research, mathematical modeling, and programming. During this phase, the students receive weekly guidance and feedback from the lecturer. Furthermore, this work phase is structured through specific milestones, e.g., discussing a preliminary outline, first results, and main findings. Finally, the students create a final paper on their work and presentations as well as discussions of all papers take place such that the students reflect their work and train their presentation skills.

Course Criteria & Registration

When you register for the course, the system will put you on the wait-list. Additionally, please send your CV and transcript to tobias.enders@tum.de. Your transcript must show the grade from the lecture "Introduction to Deep Reinforcement Learning". If you took the exam for the lecture in February 2023 and the grade is not shown on your transcript yet, please state that in your e-mail (you can of course still register in this case). After the registration deadline, in the week of April 10, we will review your documents and assign the 21 available spots based on your application.

Syllabus

• Modeling and Optimization in Operations Management (or comparable course)

• Good knowledge in programming and modeling with Python / Gurobi (or comparable programming experience)

• Recommended: Good knowledge in a high-level programming language (e.g., C++)

Syllabus

Transportation systems are seen as some of the most pervasive and influential systems in any society or economy as they are key enabler for central achievements, e.g., individual mobility, trade, globalization, and wealth. Accordingly, designing such systems in a sustainable way remains a key challenge for today’s cities and researchers. Here, a variety of new concepts evolved, e.g., autonomous mobility on demand systems for urban passenger transport, the electrification of (urban) logistics fleets, smart grids that capture the interplay between electric vehicle charging and the power network, micro hub networks in city logistics, intermodal and multimodal transport, mobility as a service applications.
In this seminar, we cover recent enhancements in these fields and learn how to implement large-scale algorithms that are amenable for application in practice.

Timeline [preliminary]

Participation in the lectures and presentations is compulsory. The lectures and presentations will be held in person on the TUM campus in Munich.

Q&A sessions via Zoom or in person on the dates without lectures.

Topic

You will be working on a case-study with a focus on sustainable transportation systems. We will present and discuss the case study in the first two sessions.

Grading

The grading bases on the intermediate presentation (10%), the final presentation (25%) and the written report (65%). 

Enrollment

This course is both in high demand and challenging; if you want to participate in this course, make sure that you do not miss the following steps:

1. Enroll via TUMonline during the application period.

2. Please send us a brief application until the end of the application period, including a CV and your transcript of records, indicating that you meet the mandatory prerequisites to lehre.bais@mgt.tum.de.

3. After the assignments are carried out, if you wish to deregister, send a mail to seminars@mgt.tum.de including the deregistration form . Only then, you can be assigned a fixed place in another seminar.

4. To stay enrolled, you have to attend the first lecture. If you got assigned to our waiting list, we recommend that you also attend the first lecture to move up if somebody else does not attend.