DSS
Decision Sciences & Systems
Technical University of Munich
 

If you are interested in a particular topic listed here for a Bachelor or Master thesis, please contact the corresponding person from the list below. If you are interested in writing a thesis on another (non-listed) topic within the scope of our group or you want to participate in guided research or an interdisciplinary project, write an email to Felipe Maldonado. Please state your skills and interests and also attach a current CV and a recent grade report. First contact should be established at least one month before registration of the project in order to allow for sufficient time to settle for a suitable topic.

 

TitleFocusContact
Optimization and Market Design
(BSc or MSc thesis)
various topics

Prof. Martin Bichler

Computational Social Choice and Algorithmic Game Theory various topics (having attended one of the courses "Computational Social Choice" or "Algorithmic Game Theory" or seminars "Multiagent Systems" or "Economics & Computation" is recommended)

Prof. Felix Brandt

Sample Efficiency in Multiagent Reinforcement Learning via Quasirandomness

 

(MSc Thesis or IDP/Guided Research, etc.)

Our current research explores the computation of market equilibria in game-theoretic settings via multi-agent machine learning techniques: To do so, market participants update their behavior according to data of past market outcomes. To achieve good precision, this method requires the simulation of millions of episodes and computationally intensive Monte-Carlo integration. 
In this thesis, you will focus on methods from numerical analysis (e.g. stratified sampling, quasirandom numbers, variance reduction techniques) to analyze and improve sample-efficiency in this setting. The thesis should comprise both theoretical analysis and empirical work using implementations in python/pytorch.

This thesis can be supervised for students in the MA or IN departments.

Required: Strong background in mathematics, esp. probability theory, and numerics. Previous experience with programming in python. Basic understanding of game theory and machine learning.

Stefan Heidekrüger

Cooperative Multi-Agent Reinforcement in the Flatland Environment

(MSc Thesis)

In this thesis you will perform MARL in the Flatland environment (https://www.aicrowd.com/challenges/neurips-2020-flatland-challenge/) was/is part of a recent NeurIPS competition. You'll research and reproduce existing state-of-the-art solutions and implement your own extensions.

Required: Some (demonstrated) prior experience with Reinforcement Learning, in-depth experience with Deep Learning, strong programming skills in python and tensorflow and/or pytorch (or another deep learning framework).

Stefan Heidekrüger
Game Theoretic Analysis of Election Campaign Spending

(BSc Thesis, MSc possible with extended scope)

In this thesis you will explore how the budget-allocation problem of political campaigns can be modeled as a Game (e.g. generalized Colonel-Blotto games) and collect and analyze data from recent elections (e.g. 2020 U.S. general election) to investigate whether actual campaign spending corresponds to game-theoretic equilibrium strategies.

Required background: Game Theory (e.g. course on Algorithmic Game Theory IN2239), hands-on experience working with data (e.g. SQL / R / python+pandas)

Stefan Heidekrüger
Solving PDEs with Deep Neural Networks

In Game Theory the established solution concept of Bayes Nash equilibria can sometimes be stated as a differential equation in certain situations. In this project, these equations will be analyzed and solved numerically via standard methods and procedures based on Neural Networks.

Mandatory Requirements: Basic knowledge of differential equations and numerical methods. Optional Requirements: Python, Deep Learning, RL, PyTorch.

This thesis can be supervised for students in the MA or IN departments.

Literature: J. Sirignano and K. Spiliopoulos, “DGM: A deep learning algorithm for solving partial differential equations”, 2018.

Nils Kohring
Simulations and analysis in shared-economy markets

The sharing economy depends on the development of the sharing platform. Different platforms (e.g., ride-hailing, freight exchange, kidney exchange, resource allocation, ... ) have different characteristics. We are committed to abstracting mathematical models from reality to simulate, analyze and provide theory. Research issues include but are not limited to matching strategies, pricing issues, and online prediction.

Required: advanced programming skills (e.g., Python, Matlab, at least one), mathematics, operation research.

Donghao Zhu

Modifying a bid language for
spectrum auctions

(BA Thesis)

The Flexible Use and Efficient Licensing (FUEL) bid language was proposed for conducting a radio spectrum auction in the US to allocate licenses for the new 5G network to telecommunication providers. In this thesis it shall be analyzed how small modifications of this bid language influence the runtime and efficiency of the allocation problem.

required: advanced C++ skills, helpful: knowledge in auction theory & operations research

Gregor Schwarz
Simulation of airport time slot auctions

We provide a simulation system that can be extended by further valuation models for airlines and/or bidding languages. In simulation experiments, the impact on prices, the welfare distribution, and computational costs should be analyzed for different payment rules.

required: advanced programming skills (python); helpful: auction theory, operations research

Paul Karänke

Development of a Simplex Algorithm Visualizer

(BA Thesis)

The teaching staff of our Operations Research lecture would like to provide students with a set of tools that allow them to practice various parts of the curriculum online. For this purpose, we aim to develop a visualization tool for the Simplex Algorithm. It should work with linear programming problems in 3 variables, so usage of some 3D-visualization library is required. A basic version of this tool should visualize the feasible region. At each step of the algorithm, it should indicate, along which edge the Simplex algorithm improves the solution, and which change of basic variables corresponds to this in the Simplex tableau. The tool should be written in a programming language that allows usage in a web browser (probably Javascript together with three.js).

Requirements:

  • Deep knowledge of the Simplex Algorithm.
  • Programming experience in Javascript.
  • Preferably, but not required, some knowledge in 3D-visualization.

Maximilian Fichtl,

Gregor Schwarz

Development of a Sensitivity Analysis Tool 

(BA Thesis)

The teaching staff of our Operations Research lecture would like to provide students with a set of tools that allow them to practice various parts of the curriculum online. For this purpose, we aim to develop a tool that generates (simple) Simplex Algorithm instances and then automatically performs a sensitivity analysis on them. The results of the sensitivity analysis should then be outputted in a generic output format. In addition to that, the tool should also be capacble of generating proper problem questions and parse them to a Latex document.

Requirements:

  • Deep knowledge of the Simplex Algorithm.
  • Programming experience in Java.

Maximilian Fichtl,

Gregor Schwarz

 

Visualization of Flow Algorithms with TikZ

(BA Thesis)

The teaching stuff of our Operations Research lecture would like to provide students with a set of tools that allow them to practice various parts of the curriculum online. For this purpose, we aim to develop a tool that generates simple flow networks. Different maximum flow algorithms (Ford-Fulkerson, Edmonds-Karp, Dinic) should then be implemented and executed on the previously generated flow networks. In each iteration, the tool should visualize the current flow network, the residual graph, and the augmenting path with TikZ.

Requirements:

  • Deep knowledge of various flow algorithms.
  • Programming experience in Java.
  • Helpful: First experiences with TikZ

Maximilian Fichtl,

Gregor Schwarz

 

Templates and Information for Creating Theses:

Thesis Template (latex)

Slides Template (ppt & latex)

General Information for Theses

 

Decision Sciences & Systems (DSS), Department of Informatics (I18), Technische Universität München, Boltzmannstr. 3, 85748 Garching, Germany
©2002-2020 DSS All Rights Reserved
Impressum, Privacy Policy, Copyright Information and Disclaimer

We use cookies on our website. Some of them are essential for the operation of the site, while others help us to improve this site and the user experience (tracking cookies). You can decide for yourself whether you want to allow cookies or not. Please note that if you reject them, you may not be able to use all the functionalities of the site.

Ok