Tomek Korbak bio photo

Tomek Korbak

machine learning,
cognitive science,

Email Twitter Scholar LinkedIn GitHub

I’m interested in machine learning, cognitive science and philosophy of science.

Currently, I’m the principal investigator in the project “Formal and computational models of self-organization in cognitive science” at the Institute of Philosophy and Sociology, Polish Academy of Sciences. The project is supervised by Prof. Marcin Miłkowski and funded by Polish Ministry of Science and Higher Education. The goal is to develop an integrative account of explanation employing the concept of self-organization, mostly focusing on Bayesian approaches in computational neuroscience, representation learning in deep neural networks and enactive cognitive science.

I’m also collaborating with Human Interactivity and Language Lab at the Faculty of Psychology, University of Warsaw on the project “Developmentally informed agent-based modeling of symbolic constraints in interaction” led by Prof. Joanna Rączaszek-Leonardi. My contribution is investigating the emergence of compositional communication in multi-agent systems.

I studied cognitive science, philosophy and physics at the University of Warsaw within an interdepartmental individual course of studies (MISH). I’m currently finishing my master’s degree in cognitive science. I’m a fellow of Collegium Invisibile and a recipient of the Minister of Science and Higher Education scholarship for exceptional students in 2016.

Apart from academia, I also work as a machine learning engineer at Sigmoidal, focusing on natural language processing.


Our paper, Developmentally motivated emergence of compositional communication via template transfer, was accepted to NeurIPS 2019 workshop Emergent Communication: Towards Natural Language.