Principal Investigator: Karina Gibert
Email: karina.gibert@upc.edu
Web: https://ideai.upc.edu/en ; https://eio.upc.edu/en/homepages/karina
Brief Theme Description:
The selected candidate will contribute to the development of innovative methodologies for explainable AI, specifically within the context of hybrid intelligent decision support systems. These systems integrate knowledge-based components with data-driven methods to enhance decision-making processes.
The main goal of the research is to bridge the gap between effective decision-making and the results of data-driven models, sometimes non-trustable by decision-makers. The generation of explainability elements around data-driven methods in both supervised and unsupervised machine learning methods is a long-term research line of the hosting group and the current research in explainable AI is pushing the work of the group in this topic.
Additionally, the research may explore the inherent tension between explainability, transparency, and concerns such as privacy or statistical secrecy, presenting an open challenge that warrants attention.
The application domains for this work may range across various fields, including water management, health and wellness, and industrial applications. This variety underscores the practical impact and interdisciplinary nature of the research.
Available Infrastructures: UPC Cloud and Computing Resources.
Possible Secondments: Several possibilities in Europe or UK.
Keywords: Explainable AI; Hybrid AI; Intelligent Decision Support Systems; Ethics in AI; Transparency; Privacy.
