Incontro INdAM on

Mathematics of Machine Learning

In recent years, machine learning algorithms have seen unprecedented success in a variety of models, ranging from science and engineering to medicine and social sciences. As a result, artificial intelligence has been identified as one of the key technologies for future social and economic advance. However, as such algorithms are increasingly used in sensitive areas such as medical devices and autonomous driving, there is a growing need for a systematic understanding of their operating principles and for guarantees on their accuracy and reliability. To address these challenges, there has been a significant increase in interest and efforts to develop a mathematical theory of machine learning models. This interdisciplinary endeavor critically relies on communication between specialists from various fields of mathematics, as well as empirical research.

This conference aims to bring together experts in various mathematical aspects of the theory of machine and deep learning, fostering discussions and collaborations that will advance our understanding of the mathematical principles underlying these models. More specifically, we plan to attract young researchers and world-leading scientists with backgrounds in

ˆ Dynamical Systems Theory

ˆ High-dimensional Statistics

ˆ Mathematical Analysis

ˆ Probability and Stochastic Processes

ˆ Statistical Physics

ˆ Theory of Optimization

in an effort to promote new cross-disciplinary collaborations by sharing re- cent results and open problems in the flourishing domain of the mathematical foundations of machine learning.