# Markov Blankets
- https://royalsocietypublishing.org/doi/10.1098/rsif.2017.0792
- In [statistics](https://en.wikipedia.org/wiki/Statistics) and [machine learning](https://en.wikipedia.org/wiki/Machine_learning), the **Markov blanket** for a [node](https://en.wikipedia.org/wiki/Vertex_(graph_theory)) in a [graphical model](https://en.wikipedia.org/wiki/Graphical_model) contains all the variables that shield the node from the rest of the network. This means that the Markov blanket of a node is the only knowledge needed to predict the behavior of that node and its children. The term was coined by [Judea Pearl](https://en.wikipedia.org/wiki/Judea_Pearl) in 1988.[[1]](https://en.wikipedia.org/wiki/Markov_blanket#cite_note-1)