Talks and Public
Value System Aggregation for ethical decision-making.
This talk explores a novel approach to addressing the challenge of aligning decisions with human moral values, particularly in the context of AI systems. Central to this research is the idea of creating a shared value system that can represent diverse individual preferences while minimizing disagreements in decisions. We introduce a new concept, Minimal Decision Divergence, which ensures that collective decisions are as close as possible to those made by individuals based on their unique value systems. Building on this principle, we propose and analyze a mathematically sound aggregation method designed to support value-aligned decision-making. Through real-world data and examples, we demonstrate how this approach can be applied effectively, highlighting its implications for ethical AI and decision-making in multi-agent systems. The talk aims to make these concepts accessible and relevant, showcasing the importance of principled methods for aggregating values in practical scenarios.
Scientific Monologue on Dirichlet's principle
Participated in the spanish-speaking science divulgation competition called CienciaShow and organised by the BigVan group with a monologue on funny monologue on mathematics and the dirchlet principle.