Two scientific articles by Cenia researcher Yamisleydi Salgueiro were published by leading international journals on artificial intelligence: IEEE Transactions on Cybernetics and Neural Computing and Applications. Yamisleydi Salgueiro is an expert in computer science and artificial intelligence, with her main areas of expertise being multi-objective optimization, deep learning and smart microgrids. Currently, she is an academic at the Faculty of Engineering of the University of Talca and at Cenia she participates in the research line “Deep Learning for Vision and Language” (RL1).
The paper entitled Long Short-term Cognitive Networks, written by the researcher together with Gonzalo Nápoles, Isel Grau and Agnieszka Jastrzebska, was published in the journal Neural Computing and Applications. This journal is dedicated to collecting original research in the field of practical applications of neural computing and related techniques such as genetic algorithms, fuzzy logic and fuzzy systems.
The second paper, entitled Recurrence-Aware Long-Term Cognitive Network for Explainable Pattern Classification, was carried out in collaboration with Gonzalo Nápoles, Isel Grau and Maikel León Espinosa, and selected for publication by the prestigious IEEE Transactions on Cybernetics, which is in the top 10 journals of the Web of Science in the Artificial Intelligence category: “The articles published there have a wide reach among the academic community, which allows us to increase both the visibility of our research and its impact,” says the researcher.
Salgueiro believes that these projects are relevant for two main reasons. On the one hand, both papers present neural systems where novel learning architectures and algorithms are articulated, and on the other hand, they address mechanisms that allow the interpretability, to a certain extent, of the models generated. The latter is important, since it is related to the values of transparency and reliability that are currently expected in the development of artificial intelligence technologies.
Check out the original publications here:
Neural Computing and Applications
IEEE Transactions on Cybernetics