MIT Press

Table 1:

Mathematical Notation.

Dimensions (number of features)

Data

⁠, point n, feature i

Masks

Cluster label

Total number of clusters

Mixture weight, cluster mean, covariance

Probability density function of the multivariate gaussian distribution

Total number of data points

Number of points for which feature i is masked

Noise mean for feature i

Noise variance for feature i

Virtual features (random variable)

Mean of virtual feature

Variance of virtual feature

Log likelihood of

in cluster k

Set of data points assigned to cluster k

Subset of

for which feature i is fully masked