What Are Markov Blankets?

In a Bayesian Network, the Markov Blanket of a variable is the minimal set of variables that provides all the necessary information to predict the value of that variable. Once you know the variables in a Markov Blanket, the rest of the network doesn't contribute additional predictive power.

The Markov Blanket consists of:

1. Parents: The variables that directly influence the variable in question.
2. Children: The variables that are directly influenced by it.
3. Co-parents: Other variables that influence the children of the variable.

In essence, the Markov Blanket isolates a variable from the rest of the network, allowing for more efficient predictions and computations.

Read more:

The Markov blankets of life: autonomy, active inference and the free energy principle