Towards Causal Representation Learning
20 October 2022 at 13:00 (UK Time)
By: Francesco Locatello
Amazon Web Services (AWS), Tubingen, Germany
Nowadays there is strong cross-pollination between the machine learning and graphical causality fields, with increasing mutual interest to benefit from the respective advances. In this talk, I will first review fundamental concepts of causal inference and present new approaches for causal discovery using machine learning. Second, I will broadly discuss how causality can contribute to modern machine learning research. Third, I will introduce causal representation learning as an open problem for both communities: the discovery of high-level causal variables from low-level observations. Finally, I will discuss my work on learning (more) causal representations, the architectural innovations that are required to represent causal variables with neural networks, and the benefits of leveraging causal principles as inductive biases in deep learning.
Bio: Dr Francesco Locatello is a Senior Applied Scientist at Amazon AWS where he leads the Causal Representation Learning team. He obtained his PhD at ETH Zurich supervised by Gunnar Rätsch (ETH Zurich) and Bernhard Schölkopf (Max Planck Institute for Intelligent Systems). He held doctoral fellowships at the Max Planck ETH Center for Learning Systems, ELLIS, and received the Google PhD Fellowship in Machine Learning 2019. His research has won several awards, including a best paper award at ICML 2019 and the ETH medal for outstanding doctoral dissertation.
Full details can be found on the Mind and Machine website https://mindandmachine.blogs.bristol.ac.uk/seminars/