The training school will be held at the University of Coimbra, Portugal, during 2 full days in January 2026. We aim at gathering together PhD candidates and young researchers with highly qualified trainers, mostly from the area of machine learning (ML) as applied to computational chemistry problems. The central topic of the school is the parameterization of force fields to be used in molecular simulations by employing either classical fitting approaches or recently proposed ML techniques. Lectures covering introductory ML concepts and applications to potential energy modeling will be delivered by experts. Although the subject will be considered in a broad context, particular emphasis will be given to applications regarding force fields parameterization with interest in biological systems and materials. Additionally, practical and tutorial classes that complement the theoretical lectures will be delivered by experts with the help of younger researchers with expertise in the area. The practical activities will allow trainees to learn how to use public domain software for the parameterization of force fields.

The COST Action COSY aims to provide a computationally and experimentally sound foundation for the fundamental understanding and control of confined molecular systems. The resulting outcome will be translated into useful knowledge forming the basis for applications. These range from creating a new generation of materials including biomaterials, with immediate transfer to industry, to disclosing the chemistry occurring in space. To this end, COSY will combine new cutting-edge experimental techniques for the synthesis of novel nanomaterials and high-resolution characterization thereof, with state-of-the-art first principles modelling. The most advanced methods for molecular motion as well as modern artificial intelligence, machine learning technologies, and big data science will be applied.

COSY COST Action coordinators: María Pilar de Lara-Castells (Chair), Cristina Puzzarini (Vice Chair)