Program
Note that all talks will be 20 minutes (Q&A included).
14–14:45 INVITED TALK. TERESA SCANTAMBURLO: Moral Exercises for Responsible AI.
Chair: Francesca Alessandra Lisi
Abstract: Artificial Intelligence (AI) scholars and practitioners have developed a vast range of tools and methodologies to address the ethical challenges raised by critical AI applications. Nevertheless, the mere availability of ethical toolkits is not sufficient to ensure responsible AI practices. In this presentation, I will introduce the idea of moral exercises to stimulate and guide meaningful conversations around choices and actions interrogating the AI actors along the whole AI-life cycle. Built upon classical and modern moral theories, moral exercises are structured activities that can offer guidance in moral reflection and judgment in AI-related tasks. In line with other approaches, such as value sensitive design, they suggest the need to rediscover the dimensions of personal commitment and intersubjectivity in responsible design practices, highlighting the value of community and dialogue. To explore the practical implications of moral exercises, I will consider critical choices in the field of algorithmic fairness and bias.
SESSION I: Explainable AI and Ethics
Chair: Melissa Antonelli
Federico Sabbatini and Roberta Calegari: Unveiling Opaque Predictors via Explainable Clustering: The CReEPy Algorithm
Miriam Doh, Caroline Mazini Rodrigues, Nicolas Boutry, Laurent Najman, Matei Mancas and Hugues Bersini: Bridging Human Concepts and Computer Vision for Explainable Face Verification
Silvana Badaloni, Antonio Rodà and Martino Scagnet: An Italian dataset for the analysis of gender stereotypes in textual documents
Andrea Apicella, Salvatore Giugliano, Francesco Isgro, Andrea Pollastro and Roberto Prevete: An XAI-based masking approach to improve classification systems
COFFEE BREAK (16:30–17)
SESSION II: Foundations and Algorithms for Ethical AI
Chair: Fabio Aurelio D'Asaro
Antonio Carnevale, Antonio Lombardi and Francesca Alessandra Lisi: Exploring Ethical and Conceptual Foundations of Human-Centred Symbiosis with Artificial Intelligence
Francesco Antonio Genco, Greta Coraglia, Giuseppe Primiero, Fabio Aurelio D'Asaro, Davide Posillipo, Davide Giannuzzi and Christian Quaggio: BRIOxAlkemy: A Bias detecting tool
Melissa Antonelli: Towards a Logical Foundation of Randomized Computation
Celeste Veronese, Daniele Meli, Filippo Bistaffa, Manel Rodríguez-Soto, Alessandro Farinelli and Juan Antonio Rodríguez-Aguilar: Inductive Logic Programming For Transparent Alignment With Multiple Moral Values
Mattia Petrolo, Ekaterina Kubyshkina and Giuseppe Primiero: A logical approach to algorithmic opacity