Towards an Epistemic Generative AI by Prof. Fabio Cuzzolin of Oxford Brookes University

The Epistemic AI is an approach which proposes the use of second-order uncertainty measures for quantifying epistemic uncertainty in artificial intelligence. A mathematical framework which generalises the concept of random variable, random sets, for instance, enable a more flexible and expressive approach to uncertainty modeling. We discuss ways in which the random sets and credal sets formalisms can model classification uncertainty over both the target and parameter spaces of a machine learning model (e.g., a neural network), outperforming Bayesian, ensemble and evidential baselines. We show how the principle can be extended to generative AI, in particular large language models less prone to hallucination, as well as diffusion processes and generative adversarial networks. Exciting applications to large concept models and visual language models as well as neural operators, scientific machine learning and neurosymbolic reasoning are discussed.

Biography:

Fabio Cuzzolin is Professor of Artificial Intelligence at Oxford Brookes University and the inaugural Director of the Institute for AI, Data Analysis and Systems (AIDAS). He received his PhD from the University of Padova with a thesis on generalized probability theory, and has held research positions at Politecnico di Milano, UCLA, and INRIA Rhône-Alpes as a Marie Curie Fellow.