Rustam Guliyev
About Me
I am a PhD student in the department of Computer Science. My research is funded by EPSRC Doctoral Training Partnership Award 2021Link opens in a new window. I am working with Prof. Hakan FerhatosmanogluLink opens in a new window who is my supervisor.
Before undertaking my PhD, in 2020, I graduated with a Computer Engineering Bachelor's degree from Middle East Technical University NCCLink opens in a new window. The following year, I founded two startups relating to international shipping and crowdsourced data gathering/labelling for AI.
Research
My research interests are broadly in graph & hypergraph-based machine learning, graph partitioning, continual learning and streaming systems. Continual learning approaches pertaining to popular ML models do not translate well into graph-based models. Furthermore, communicational overheads of distributing graph workloads bring additional challenges, especially in the case of dynamic graphs. I explore the use of modern distributed stream processing systems, such as Apache Flink, in the domain of GNNs and HGNNs to perform low-latency inference and continual learning under massive graph streams.
