Paolo Giudici: Safe Artificial Intelligence

Abstract

The growth of Artificial Intelligence applications requires to develop risk management models that can balance opportunities with risks. We contribute to the development of AI risk management models proposing a set of integrated statistical metrics that can measure the Sustainability, Accuracy, Fairness and Explainability of any Artificial Intelligence application. Our metrics are consistent with each other, as they are all derived from a common underlying statistical methodology: the Lorenz curve. Our experimental results indicate that they are easy to interpret, and that they can be applied to any machine learning method, regardless of the underlying data and model.

About

Paolo Giudici is Professor of Statistics at the University of Pavia (Italy) with over 200 Master's and 21 PhD students. Author of more than 200 publications (h-index 52) in top journals like Series B, Biometrika and Journal of Banking and Finance. His research program focuses on statistical learning, financial risk management, machine learning and safe AI and Bayesian analysis. Paolo Giucici is Coordinator of 11 funded projects, including EU Horizon 2020 projects, Chief Editor of Statistics and Artificial Intelligence in Finance and Associate Editor International Journal of data science. Fellow at the Bank for International Settlements and UCL Blockchain Center, Expert for AI Office, EIOPA and the Italian Ministry of Development.

More about the talk