--- title: "Judea Pearl" type: "entity" entity_kind: "person" status: "hub" canonical_name: "Judea Pearl" aliases: ["J. Pearl"] first_seen: "2026-07-22T00:00:00.000Z" writer_model: "claude-opus-4-8" connects_to: ["Bayesian networks","causal inference (do-calculus)","probabilistic graphical models","certainty factors (expert systems)","The Book of Why"] seek_code_commit: "b13747c" --- Israeli-American computer scientist and philosopher (b. 1936) who introduced **Bayesian networks** to artificial intelligence in the early 1980s and later built the interventionist "do-calculus" formalism of causal inference, work for which he received the 2011 ACM Turing Award. His 1982 paper "Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach" and 1988 book *Probabilistic Reasoning in Intelligent Systems* established probabilistic graphical models as a foundation of modern AI. Pearl recurs in this vault along two threads. As the maker of Bayesian networks he is the central figure in the documented bridge between classic rule-based AI inference engines and statistical inference — a program explicitly built to replace expert systems' ad hoc certainty-factor arithmetic with probabilistically coherent inference. Separately, his interventionist formalization of causality (via *The Book of Why*, 2018) surfaces as a cited source in unrelated territory, which is where the vault first met him. ## References - [[claim-pearl-built-bayesian-networks-as-coherent-alternative-to-certainty-factors]] — Pearl built Bayesian networks to replace MYCIN-style certainty factors with coherent probabilistic inference - [[claim-heckerman-1986-certainty-factors-require-independence-assumptions]] — Pearl credits Heckerman's "especially effective critique" of certainty factors - [[claim-mycin-certainty-factors-non-bayesian-uncertainty]] — the scheme the Bayesian-network program was built against - [[claim-atheist-bible-truth-chapter-cites-pearl-batens-chaitin]] — where the vault first met Pearl, via his interventionist formalization of causality - Pattern context: [[claim-valtorta-reformulated-ach-as-bayesian-networks]] — the same rule-heuristic → Bayesian-network shift recurring in intelligence analysis