Gustav Šír, Ph.D.

Gustav is an assistant professor in the Intelligent Data Analysis lab at the Department of Computer Science of the Faculty of Electrical Engineering at Czech Technical University in Prague.

He is generally involved in Artificial Intelligence research, with a particular focus on deep relational machine learning – i.e. learning from data with complex dependencies and structures such as (knowledge) graphs and databases, for which he likes to combine formal representations from mathematical logic with ideas from deep learning. Additionally, his interests include ML for predictive trading and thinking about social impacts of AI.

Graduate

Czech Technical University: Ph.D. in Artificial Intelligence & Biocybernetics;MSc in Artificial Intelligence

Undergraduate

Czech Technical University: B.Sc. in Intelligent Systems 

Representative Publications 

• Sourek, G., Aschenbrenner, V., Zelezny, F., Schockaert, S., & Kuzelka, O. (2018). Lifted relational neural networks: Efficient learning of latent relational structures. Journal of Artificial Intelligence Research, 62, 69-100.
• Sourek, G., Zelezny, F., & Kuzelka, O. (2020) Lossless Compression of Structured Convolutional Models via Lifting. In International Conference on Learning Representations.
• Sourek, G., Železný, F., & Kuželka, O. (2021). Beyond graph neural networks with lifted relational neural networks. Machine Learning, 110(7), 1695-1738.
• Hubáček, O., & Šír, G. (2023). Beating the market with a bad predictive model. International Journal of Forecasting, 39(2), 691-719.

Skills