Invited Talks

Outcomes of the FAIR NAEL project: energy-efficient and sustainable technology AI solution for pattern recognition

• Presenting author: Francesco Carlo Morabito

• Other authors: Nadia Mammone, Cosimo Ieracitano, Mario Versaci

• Affiliations: University Mediterranea of Reggio Calabria, CNR – ISASI Pozzuoli, Italy

The present speech will show the methodological approaches proposed within the PNRR FAIR NAEL project and the results achieved, specifically in healthcare and BCI. It includes a detailed description of the pre-processing of the data used to reduce the computational cost (and, thus, the energy consumption) in AI applications (i.e., classification, segmentation, and multimodal fusion). In addition, some meta-learning approaches are discussed that improve generalization performance also in different tasks, like in domain adaptation. Although the project is mainly focused in clinical applications, the talk will include some examples of the use of the same methodological approaches for cultural heritage data.