Loading timeline…
19901990s
Bayesian Networks
Probabilistic graphical models representing causality and uncertainty.
Why It Was Important
Fully developed in the 1990s by Judea Pearl, Bayesian networks represented variables and their causal dependencies as a directed acyclic graph. This allowed AI to definitively reason about probabilities, dramatically outperforming old, brittle expert systems in domains rife with uncertainty, like medical diagnosis.
Who Invented It
Judea Pearl
UCLA professor and Turing Award winner for his work on causality.
Applications
- Medical Diagnosis
- Spam Filtering
- Risk Analysis
Key Papers
- Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference
Judea Pearl · Morgan Kaufmann · 1988
Videos
Bayes' Theorem, Clearly Explained!!!!
StatQuest with Josh Starmer