11/3/2025 Seminar: Mario Chacón Falcón, ICMAT, Spain

Abstract: Artificial intelligence faces security challenges at many levels, such as the exposure of sensitive data, the vulnerability of distributed learning systems, and the need to design robust policies under adversarial uncertainty, to name but a few. In this seminar, I will discuss two approaches to improve AI security. The first focuses on federated learning, which is a branch of Machine Learning that aims to train models across distributed data sources while preserving privacy. The second part is structured around Adversarial Risk Analysis (ARA), which offers a decision-theoretic approach for defensive planning in situations of strategic uncertainty, representing attackers as stochastic agents with unknown intentions. Together, these perspectives contribute towards building more secure and reliable AI systems.

Who: Mario Chacón Falcón,  PhD candidate, ICMAT, Spain

Profile: https://mchacon28.github.io/

When: November 3, 2025, 14:00

Where: Institute of Information Theory and Automation (UTIA) in room 474

Language: English