The Science of Cause and Effect - From Deep Learning to Deep Understanding

Explore the foundations of causal reasoning and its implications for artificial intelligence in this thought-provoking lecture by Turing Award winner Judea Pearl. Delve into the concept of "understanding" in computational systems, examining the three levels of causal inference: prediction, intervention, and counterfactuals. Learn about Pearl's proposed formal definition of understanding and the computational model that facilitates reasoning at these levels. Discover how this framework can be applied to generate explanations, generalize across domains, integrate data from multiple sources, and recover from missing information. Gain insights into the future of automated scientific exploration, personalized decision-making, and social intelligence. Through practical examples and theoretical discussions, grasp the fundamental principles of causal inference and their potential to revolutionize AI and machine learning.

Introduction
Welcome
Agenda
Data vs Science
Scientific Paradigm
Practical Cases
Civil Right
Structural causal model
First law of causal inference
Counterfactuals
Ladder of causation
Law of independence
Do calculus
Estimating causal effect
Sport Medicine Example