B0894
Title: Explainable AI: Empowering machine learning models with explanations
Authors: Mattia Setzu - University of Pisa (Italy) [presenting]
Abstract: In the past years, explainable AI (XAI) has become a major field of interest for many AI researchers and practitioners, as well as domain experts and common users who leverage AI tools in their work and daily lives. The purpose is to delve into XAI with a holistic overview of XAI taxonomy, algorithms of note, advancements in the field, and its taxonomy. Different families of explanations are reviewed, their strong and weak points and some considerations are included on explanations and their impact on individual and societal use and trust of AI systems. Finally, open challenges and practical considerations for implementing XAI in real-world scenarios are concluded with.
