symmetry_correction: A function to enforce the symmetry correction after Markov...
In kelvinyangli/mbmml: Learn Markov blankets using Minimum Message Length

This function enforces the symmetry check for each learned mb. That is, if a node x is is mb(y), but y isn't in mb(x), then either add y into mb(x) or delete x from mb(y), depending on which deterministic rule to follow. The two famous deterministic rules are the Intersection and Union rules. For MB discovery using MML, we prefer to use the Intersection rule due to the high precision and low recall of mml_cpt.

1	symmetry_correction(vars, mbList, rule)

`vars`	A vector of all variables.
`mbList`	A list of learned Markov blankets.
`rule`	A string argument takes either "Intersection" or "Union", indicating the two deterministic rules mentioned above.