Master's thesis: Conditional Federated Learning for Distribution Shifts

Background. Federated Learning (FL) enables collaborative model training across decentralized data sources without sharing raw data, making it highly relevant for edge computing in emerging technologies like 6G and the Internet of Things (IoT). It supports privacy by design, as sensitive data remains on-device. However, real-world FL often struggles with heterogeneous data distributions across clients. This project addresses that challenge by exploring Conditional Federated Learning, a novel approach that combines personalized and clustered FL to better manage distribution shifts. The project is hosted by RISE Research Institutes of Sweden, a state-owned research institute that supports sustainable innovation across academia, industry, and the public sector.

Description. This thesis investigates the design and implementation of a Conditional Federated Learning framework that adapts to client-specific data distributions. The method will be tested across various tasks, model architectures, and datasets to evaluate its robustness to different types of heterogeneity. Results will be benchmarked against established FL baselines to assess performance in both accuracy and generalization.

Key Responsibilities

• Literature review on Federated Learning, personalization, and distribution shifts

• Implement Conditional Federated Learning for multiple tasks

• Evaluate the method on diverse datasets under varying degrees of heterogeneity

and compare performance with established heterogeneous FL benchmarks.

• Document findings in a scientific report

Qualifications

• Strong background in mathematics, statistics, and machine learning

• Proficiency in Python and frameworks such as PyTorch or TensorFlow

• Familiarity with Cyber Security and Federated Learning is a plus

Terms

• Scope: 30 hp, one semester full-time

• Location: Luleå (or remote with agreement)

• Start: Flexible

• Compensation: 10,000 SEK for travel, materials and the like after the project is completed and approved

Please note: You need to have a valid student visa that allows you to study in Sweden during the thesis period.

Welcome with your application

Last day of application: July 29
Contact: Rickard Brännvall (rickard.brannvall@ri.se),
Check-in questions (yes/no): 1-5 are required, 6-9 are beneficial, 10 is specifically a plus