Javad Aliakbari

About me

I am a PhD student at Chalmers University of Technology, where I have been working since February 2023 in the Communication Systems Group. I am supervised by Prof. Alexandre Graell i Amat.

I received my B.Sc. in Electrical Engineering from the University of Tehran. I then completed my M.Sc. at Sharif University of Technology, specializing in communication systems. During my master’s studies, I worked on privacy and security aspects of graph data.

Before starting my PhD, I also worked in industry as a data scientist and researcher, mainly on machine learning, signal processing, and Natural Language processing.

Research interests

My research focuses on machine learning on graph-structured data, with a particular emphasis on federated learning and privacy-preserving methods. Many real-world systems, such as social networks and communication networks, naturally form graphs and are often distributed across different data owners. In such settings, data cannot be centrally collected, which makes standard machine learning approaches unsuitable.

In my work, I study federated learning for graph neural networks, with a focus on subgraph federated learning, where each client holds a local subgraph and collaborates without sharing raw data. I develop methods based on spectral graph theory, optimization, and statistical learning, and I also analyze privacy–utility trade-offs to better understand what information can be leaked during decentralized training and how it can be controlled.