Alejandro Marrero

During his doctoral studies, the researcher was the beneficiary of a FPI research fellowship awarded by the Government of the Canary Islands. This four-year contract supported his advanced studies and culminated in his attainment of a Doctorate in Computer Engineering from the University of La Laguna (ULL). His doctoral thesis was centred on the innovative application of Evolutionary Computation techniques specifically designed for the automatic generation of sets of optimisation problem instances. This specific line of work contributes significantly to the body of knowledge within the domains of combinatorial optimisation and computational intelligence. It provides the scientific community with novel methodologies that allow for a profound analysis of the structural characteristics inherent in optimisation problems, as well as a deeper understanding of the behaviour of the algorithms used to resolve them. In recognition of his academic excellence, he was awarded the Prize for the Best Academic Record in the Master’s Degree in Computer Engineering (2018-2019) and, subsequently, the Extraordinary Doctorate Award in 2024, both granted by the University of La Laguna. He served as the lead author on multiple publications in high-impact scientific journals, including the Evolutionary Computation Journal, SoftwareX, and Mathematics. Furthermore, his work has been disseminated through proceedings at highly relevant international conferences, such as the Genetic and Evolutionary Computation Conference (GECCO) and Parallel Problem Solving from Nature (PPSN), as well as key national forums like CAEPIA and MAEB. His research focuses on the application of intelligent methods that span a wide spectrum of technologies, ranging from Evolutionary Computation and Quality Diversity algorithms to Machine Learning (ML) and Deep Learning (DL) models. These are utilised for the generation of synthetic data and the prediction of algorithmic performance based on historical results. Committed to the principles of Open Science, the results of his publications are openly available via GitHub, and he has curated three open-access datasets available on the Mendeley Data platform. In terms of scientific software development, he has successfully developed and registered frameworks that integrate data visualisation techniques with Machine Learning, enabling the extraction of interpretable knowledge regarding the solution space and algorithmic performance. A notable contribution is the software known as DIGNEA, which is available in two programming languages. This framework is presented as a tool that facilitates the generation of synthetic datasets for different optimisation domains. The results obtained have been communicated through scientific publications and conference presentations, contributing to the advancement of the state of the art in the automated design of benchmarks, as well as the understanding of computational complexity and the distribution of characteristics of optimisation problem instances. Regarding internationalisation, he completed a three-month research stay at the Edinburgh Parallel Computing Centre (EPCC), affiliated with the University of Edinburgh, an institution of recognised international prestige in high-performance computing. This stay, funded by the European HPCEuropa3 project, allowed him to acquire advanced scientific-technical capabilities in parallel computing and to collaborate with researchers of excellence, strengthening his international collaboration networks.