Por Nuno Lourenço (U. Coimbra).

Grammars are widely used by researchers of the Evolutionary Computation field, specially from the Genetic Programming (GP) community.  They were first used by Peter Whigham in the Context Free Grammar-GP(CFG-GP), but they became popular with the appearance of Grammatical Evolution (GE). In spite of being widely used by practitioners of different fields, GE is not free from drawbacks.

In this talk we will discuss recent advances on Grammar-Based Genetic Programming that aim at overcoming GE’s main issues. We will present several applications of these new methods, namely on how they can be used to evolve Deep Neural Networks, develop models to automatically perform feature engineering, how they can be used in the automatic design of Machine Learning Pipelines and how they can be used to provide insights into some problem domains.

Nuno Lourenço is an Assistant Professor at the Department of Informatics Engineering of the University of Coimbra, where he obtained his PhD in Information Science and Technology in 2016. He is the current coordinator of the Evolutionary and Complex Systems (ECOS) group, and is a member of the Centre for Informatics and Systems of University of Coimbra (CISUC) since 2009. Formerly, he was appointed as a Senior Research Officer at the University of Essex in the United Kingdom. His main research interests are in the areas Bio-Inspired Algorithms, Optimisation and Machine Learning. He is the co-creator of Structured Grammatical Evolution, Probabilistic Grammatical Evolution, and DENSER, a novel approach to automatically design Deep Artificial Neural Networks using Evolutionary Computation. He served as chair in the main conferences of the Evolutionary Computation field, namely EuroGP 2020 and 2021 as program-chair, and PPSN 2018 and EuroGP 2019 as publication chair. He is member of the Programme Committee of GECCO, PPSN, EuroGP; member of the Steering Committee of EuroGP; and executive board member of SPECIES. He has authored or co-authored more than 60 articles in journals and top conferences from the Evolutionary Computation and Artificial Intelligence areas and he has been involved as a researcher in 13 projects (national and international).

RSS Meetups are monthly gatherings of LASIGE members with research interests mainly in Software Architecture, Verification, Testing, Programming Languages, Type Systems, Logic, Concurrency, and Formal Methods.