Stratos Palias

I am a postdoc fellow in the School of Computer Science at the University of Birmingham. Previously, I obtained my PhD in Computer Science under the supervision of Prof. Ata Kabán. Before that, I received a BSc and an MSc in Statistics and Data Science at the Athens University of Economics and Business. My CV can be found here.

My research interests span Statistical Learning Theory and Information Theory. Broadly speaking, I study Machine Learning algorithms from a theoretical viewpoint, and I aim to discover geometric properties on the data, that make an algorithm learn better. I am interested in aspects such as sample complexity, i.e., how many samples we need for an efficient learning, and regularisation methods, i.e., how can we constrain the hypothesis class to prevent overfitting. My commonly used mathematical tools are borrowed from Linear Algebra and High-Dimensional Probability.

Recently, I narrowed my focus predominately on Quantum Information and Quantum Channels, working with complex-valued quantum data. I am currently investigating the capacity of channels to reliably transmit information, and how does this capacity determine the learnability of a channel.