Table 2 GMM model selection results: model comparison by AIC and BIC (N=50,000, F=4).

From: Gaussian mixture model for enhancing the quality of transmission estimation in optical networks: a machine learning approach

K

Covariance (\(\Sigma\))

Log-Likelihood

p (Params)

AIC

BIC

2

diag

-21748.26

21

43538.51

43675.37

2

full

-15412.59

41

30907.19

31174.39

3

diag

-19561.06

26

39174.12

39455.51

3

full

-13527.28

44

27142.56

27626.87

Back to article page