MaLGa Colloquia - Representing scientific data for causal inference

Title

MaLGa Colloquia - Representing scientific data for causal inference

Speaker

Francesco Locatello - Institute of Science and Technology Austria (ISTA)

Abstract

Deciphering experimental observations into structural knowledge of the world is a key component of the scientific discovery process and a longstanding challenge for AI. In this talk, I will present our recent works bridging between perception and causality. I will begin with the challenge of training predictors that are causally valid proxies of otherwise latent variables. Next, I will also show how AI models enable "looking at the data first" and discovering effects in randomized trials without supervision. I will conclude with a brief exposition as to how these ideas may be applicable beyond standard causal models as well, focusing on dynamical systems.

Bio

Francesco Locatello is a tenure-track assistant professor at the Institute of Science and Technology Austria and an AI Resident at the Chan Zuckerberg Initiative. Before, he was a senior applied scientist at Amazon Web Services. He received a PhD from ETH Zürich co-advised by Gunnar Rätsch and Bernhard Schölkopf. His research received several awards, including the ICML 2019 Best Paper award, the Hector Foundation award from the Heidelberg Academy of Science in 2023, and the Google Research Scholar Award in 2024.

When

Monday, February 9th, 16:00

Where

Room 706, UniGe DIBRIS/DIMA, Via Dodecaneso 35