ELLIS Delft Talk by Hadi Jamali-Rad: Few-Shot Learning at TU Delft, Foundational AI at Shell

Abstract

Learning quickly from very few labeled samples (aka, few-shot learning) is a fundamental attribute that separates machines and humans in the era of deep representation learning. Unsupervised few-shot learning aspires to bridge this gap by discarding the reliance on annotations at training time. We will cover some of our most recent works in the realm of few-shot learning within the context of supervised, unsupervised, and generative AI. Next to that, we'll briefly reflect on some of Shell's initiatives focused on Al-enabled renewable power generation and low-carbon fuels (solar, wind, e-mobility, nature based solutions and biotechnology) to tackle climate change and to become net-zero emissions energy business by 2050.

This meeting is open for all interested researchers, and we particularly want to emphasize that we very much welcome all the PhD students and postdocs that are associated with the unit! If you have ideas about ways to make these talks more engaging for you, please let us know your suggestions.