Note: The following list is by no means exhaustive or complete. There is always some student work to be done in various research projects, and many of these projects are not listed here. Don’t hesitate to drop an email to any member of the chair asking for currently available topics in their field of research.
Instructions
For applying, please send the listed contact person your CV and grade report.
We usually require a thesis proposal from your side on the decided topic in the same format as the final thesis.
This allows for us to understand that you have clearly understood the topic and your writing style.
Additionally, this provides a list of tasks defined and agreed by both the involved parties ( you and us).
This proposal is in the same Format as the final thesis, thus you will already have a start when writing the thesis in the end.
We use TUM Overleaf for synchronising the written thesis.
The proposal should be of around 4-5 pages and should have at least below sections:
Introduction: Describing a bit of background, motivation and overall problem statement.
Research Questions (if any): The research questions which will be answered as part of the thesis.
Objectives: Objectives or goals which will be completed as part of the thesis.
Literature Review: Some existing work related to the problem to be tackled in the thesis.
Proposed Approach: How you will solve the problem, a high level overview of the design should be described here.
We are flexible to adapt the objectives/goals of the thesis over the course of duration of the thesis depending on the work and results.
Topics
Background
Federated learning (FL) enables resource-constrained edge devices to learn a shared Machine Learning (ML) or Deep Neural Network (DNN) model, while keeping the training data local and providing privacy, security, and economic benefits. However, building a shared model for heterogeneous devices such as resource-constrained edge and cloud makes the efficient management of FL-clients challenging. Furthermore, with the rapid growth of FL-clients, the scaling of FL training process is also difficult.
Goals
The aim of this work is to develop and implement a framework for federated learning on heterogeneous devices using FaaS based functions. We already have a baseline implementation which needs to be extended.
Ensuring fault tolerance and scalability of different components in the FL system.
Incorporating privacy and security in the framework.
Requirements
Good knowledge of ML and Deep learning.
Good Knowledge of Python.
Knowledge of FaaS platforms.
Contact
Background
Serverless computing, with function-as-a-service (FaaS) is an attractive cloud model in which the user is not responsible for server deployment and infrastructure management, but only for writing the code and packaging it. In FaaS, an application is decomposed into simple, standalone functions that are deployed to a serverless platform for execution. Although originally designed for cloud enviornments serverless computing is gaining traction in Edge Computing. To avoid dependency with a specific vendor severeral open-source serverless frameworks have been proposed. However, their usage and viability on heterogeneous edge devices is still unclear.
Goals
The aim of this work is to evaluate and analyze the performance of different open source serverless frameworks on different heterogeneous edge devices such as FPGAs/Rasberry Pis etc. for particular usecases eg. Edge AI.
Requirements
Knowledge of Docker, K8s, monitoring stack such as Prometheus
