Causal Representation Learning - Tutorial

Explore the emerging field of Causal Representation Learning (CRL) in this comprehensive tutorial presented at the Uncertainty in Artificial Intelligence conference. Delve into the innovative approach that bridges the gap between causality and machine learning, addressing how to learn causal models and mechanisms without direct measurements of all variables. Gain insights from experts Dhanya Sridhar and Jason Hartford as they guide you through the core technical problems and assumptions driving CRL. Discover how this cutting-edge research area combines recent advances in machine learning with new assumptions to identify causal variables from low-level observations such as text, images, or biological measurements. Develop strong intuitions about the fundamental concepts underpinning CRL and understand the connections across different results. Conclude the tutorial by examining open questions and potential applications of CRL in scientific discovery, providing a forward-looking perspective on this exciting field.