Students

Are you a student interested in doing a master thesis on deep learning? Feel free to contact me to discuss potential ideas. Some topic suggestions are:

• Federated and decentralized learning on non-iid data
• Out of distribution generalization
• Deep learning for aerial and satellite data

Supervised master theses

• [2023] Emilie Klefbom and Marcus Toftås Decentralized Deep Learning under Distributed Concept Drift
• [2022] Karin Bergdahl Impact of model architecture and data distribution on self-supervised federated learning
• [2021] Noa Onoszko and Gustav Karlsson PENS: Leveraging Data Heterogeneity in Federated Learning
• [2020] David Ericsson and Adam Östberg Adversarial representation learning for private speech generation
• [2018] Henrik Arnelid Sensor modelling with recurrent conditional GANs