This is the : The Akaike Information Criterion. In this model, the utility function is given by:

    AIC = m * log( Var(e) / m ) + 2 * n

    where

    Var(e) = variance of excess return of manager over benchmark, using current subset
    n = number of indices in current subset
    m = number of returns

Again, StyleADVISOR chooses the subset of indices where the utility function assumes its minimal value.