Judea Pearl is an Israeli-American computer scientist and philosopher, renowned for his advocacy of the probabilistic approach to artificial intelligence and the creation of Bayesian networks. His work has profoundly reshaped our understanding of causality, learning, and reasoning under uncertainty. Pearl introduced a paradigm shift in AI, emphasizing the comprehension of causal relationships rather than mere correlation. His innovative methodologies and theoretical contributions continue to influence the field of artificial intelligence and our grasp of complex systems.
Many of the concepts and terminology surrounding modern causal inference can be quite intimidating to the novice. Judea Pearl presents a book ideal for beginners in statistics, providing a comprehensive introduction to the field of causality.
Models, Reasoning and Inference. Ausgezeichnet: ACM Turing Award for Transforming Artificial Intelligence 2011
The book delves into the evolution of causation from a vague concept to a robust mathematical theory, highlighting its applications across various disciplines such as statistics, AI, and economics. Judea Pearl integrates different approaches to causation, providing accessible mathematical tools for exploring causal relationships and statistical associations. This revised edition addresses complex issues and recent advancements, making it valuable for students and professionals alike. Pearl's significant contributions to AI research are also recognized, enhancing the book's credibility in the field.
This paper summarizes recent advances in causal inference and highlights the necessary shifts from traditional statistical analysis to causal analysis of multivariate data. It emphasizes the foundational assumptions of causal inferences, the language used to express these assumptions, and the conditional nature of causal and counterfactual claims, along with the methods developed to assess them. The discussion is grounded in a general theory of causation based on the Structural Causal Model (SCM), which integrates various approaches to causation and offers a coherent mathematical framework for analyzing causes and counterfactuals. The paper explores mathematical tools for addressing three types of causal queries: (1) the effects of potential interventions (causal effects or policy evaluation), (2) probabilities of counterfactuals (including "regret," "attribution," and "causes of effects"), and (3) direct and indirect effects (mediation). Additionally, it defines the formal and conceptual relationships between structural and potential-outcome frameworks, presenting tools for a combined analysis that leverages the strengths of both. These tools are illustrated through analyses of mediation, causes of effects, and probabilities of causation.
How the study of causality revolutionized science and the world Cause and effect: it's at the center of scientific inquiry, and yet for decades scientists had no way of answering simple questions, such as whether smoking causes cancer. In The Book of Why, Judea Pearl and Dana Mackenzie show how Pearl's work on causality has broken through this stalemate, unleashing a revolution in our knowledge of the world. Anyone who wants to understand how science, the human mind, or artificial intelligence works needs The Book of Why. "Illuminating. . . a valuable lesson on the history of ideas." --New York Times "This book really gets you thinking about cause and effect as it applies to issues of our time. . . . Extraordinary." --Science Friday