On March 29th Blaž Škrlj received the Jožef Stefan Golden Emblem Prize for the impact of his doctoral thesis Efficient Neuro-Symbolic Machine Learning. In his work, he discussed new artificial intelligence methods that, by combining neural and symbolic machine learning methods, ensure the classification accuracy of modern neural networks and the symbolic interpretation of learned neural models. He developed new powerful neuro-symbolic learning methods that are adapted for the analysis of various types of data – from text and tabular to relational and network data – which is key to their effective use in biology, medicine, the financial industry and other sciences.

The most notable results of his doctoral thesis include a new method for efficient learning of vector embeddings of nodes on larger networks and a method for ranking features using neural networks with attention, used for detecting plant pathogens and for modeling financial flows. The work is also distinguished by its commitment to open science and the development of open-source software and web tools that can be used by the wider scientific community. Its RaKUn tool for automatic keyword search is used in the search engine of the National Library of Finland, and its CaNDis web service provides researchers with access to biological databases useful for finding causal links between diseases and pathogens.

The recording of the entire ceremony is available on: https://video.arnes.si/watch/ncv5gcjzkchv

(Blaž recieved the award on 33:43)

Congratulations!