CMI

Conditional Mutual Information (CMI) is a measure used in information theory to quantify the amount of information shared between two variables, given the knowledge of a third variable. Essentially, it tells us how much knowing one variable reduces uncertainty about another, when we already know the third. CMI is useful for understanding relationships in complex systems, helping to identify whether two variables are directly related or if their association is explained by a third factor. It’s a key tool in fields like machine learning and data analysis for feature selection and dependency assessment.