• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

MASNA teacher Nada Lavrač received the Zois Award for outstanding scientific achievements

Every year, the Government of the Republic of Slovenia awards scientists for outstanding scientific achievements. This year, Nada Lavrač, MASNA teacher for the course ‘Statistical Learning Theory’, received the Zois award for original methods of data representation for effective machine learning.

MASNA teacher Nada Lavrač received the Zois Award for outstanding scientific achievements

Nada Lavrač, former head of the Department of Knowledge Technologies of the "Jožef Stefan" Institute and professor at the Jožef Stefan International Postgraduate School and at the University of Nova Gorica, is a top expert in the field of machine learning from tabular, relational and textual data and applications of artificial intelligence in medicine and bioinformatics. Her research focuses on developing methods for learning from complex relational and network data. Prof. Lavrač is the founder of the field of symbolic machine learning of representations, called propositionalization, which enables the more efficient use of machine learning algorithms through the automated conversion of relational data into a simpler tabular form. In her recently published monograph, Representation Learning: Propositionalization and Embeddings, the cutting-edge scientific achievement of an original machine learning method for learning representations of complex relational data, which she developed with colleagues. The use of the proposed neuro-symbolic approaches to learning representations ensures the comprehensibility, expressive and explanatory power of symbolic learning models, as well as the accuracy and efficiency of modern neural networks. In addition to the description of the original methods, the monograph also provides a comprehensive overview of the research field of representation learning, achieved by establishing a common framework and uniform terminology, which enabled a unified description of representation learning techniques for the needs of classical machine learning and deep neural networks.

The award ceremony took place on December 19, 2022 in Cankarjev dom, Ljubljana.

We congratulate Nada and are happy to see her among our teaching staff.

 
Photo: Boštjan Podlogar/STA