Back to TUM-PREP Projects overview

Project Overview

Project Code: CIT 07

Project name:

Privacy, Security and Communication Efficiency in Federated Learning

TUM Department:

CIT - Electrical and Computer Engineering

TUM Chair / Institute:

Chair of Communications Engineering

Research area:

Federated Learning

Student background:

Computer Science/ InformaticsElectrical EngineeringMathematics

Further disciplines:

Participation also possible online only:

Planned project location:

Technical University of Munich
Theresienstrasse 90
80333 München

Project Supervisor - Contact Details


Title:

Given name:

Yue

Family name:

Xia

E-mail:

yue1.xia@tum.de

Phone:

015226565973

Additional Project Supervisor - Contact Details


Title:

Dr

Given name:

Rawad

Family name:

Bitar

E-mail:

rawad.bitar1@gmail.com

Phone:

Additional Project Supervisor - Contact Details


Title:

Given name:

Family name:

E-mail:

Phone:

Project Description


Project description:

Privacy, Security, and Communication Efficiency in Federated Learning
As our world becomes more data-driven, sensitive information is increasingly stored and processed across personal devices (e.g., smartphones, wearables, and IoT sensors). Federated Learning (FL) offers a powerful way to train machine learning models collaboratively without centralizing the raw data, making it a hot topic in both academia and industry.
But FL is far from a solved problem. In its standard form, FL can still leak private information through model updates, raising ethical and legal concerns. Strengthening privacy protections often introduces trade-offs, such as higher communication costs between clients and servers. On top of that, FL systems must also be secure against malicious participants, which can conflict with privacy requirements.
This project tackles the core challenge of balancing privacy, security, and communication efficiency in federated learning. The student will:
• Perform a comprehensive literature review of state-of-the-art privacy and security mechanisms in FL.
• Analyze the inherent trade-offs between privacy guarantees, robustness to attacks, and communication overhead.
• Explore and propose ideas (conceptual or practical) that can push the boundary toward efficient, secure, and privacy-preserving FL.
This project is an exciting opportunity for students interested in machine learning, security and privacy, and distributed systems. It is especially suitable for those who want to work on problems at the intersection of theory and practice, with potential real-world applications in healthcare, finance, and edge computing.

Working hours per week planned:

40

Prerequisites


Required study level minimum (at time of TUM PREP project start):

3 years of bachelor studies completed

Subject related:

Other:

Knowledge of Python programming

  • Keine Stichwörter