Skip to Main Content
Table 12 
Analysis of symmetric re-weighting–based refinement applied to the same initial mappings of our adversarial autoencoder on the Dinu-Artetxe data set.
 En-EsEn-ItEn-DeEn-Fi
→←→←→←→←
Symmetric re-weighting 46.5 42.4 37.5 31.9 48.3 44.1 32.4 32.7 
OLS 41.8 36.7 29.0 29.6 41.5 39.3 27.7 26.9 
OLS + Orthogonality 45.1 39.5 35.7 31.7 47.6 43.8 30.9 32.3 
OLS with L1 regularizer (LASSO) 42.3 36.9 28.4 30.4 42.3 39.7 26.8 27.1 
LASSO + Orthogonality 46.0 40.1 34.8 32.2 47.1 44.2 32.4 32.6 
OLS with L2 regularizer (RIDGE) 41.7 36.5 29.4 29.5 41.9 39.0 28.2 29.4 
RIDGE + Orthogonality 46.2 39.9 35.1 32.0 47.8 44.2 32.7 32.7 
OLS with L1 & L2 regularizers (E-NET) 42.1 36.8 29.3 30.1 42.2 38.2 28.5 28.5 
E-NET + Orthogonality 45.6 40.3 35.3 31.6 47.7 44.0 32.5 32.9 
 En-EsEn-ItEn-DeEn-Fi
→←→←→←→←
Symmetric re-weighting 46.5 42.4 37.5 31.9 48.3 44.1 32.4 32.7 
OLS 41.8 36.7 29.0 29.6 41.5 39.3 27.7 26.9 
OLS + Orthogonality 45.1 39.5 35.7 31.7 47.6 43.8 30.9 32.3 
OLS with L1 regularizer (LASSO) 42.3 36.9 28.4 30.4 42.3 39.7 26.8 27.1 
LASSO + Orthogonality 46.0 40.1 34.8 32.2 47.1 44.2 32.4 32.6 
OLS with L2 regularizer (RIDGE) 41.7 36.5 29.4 29.5 41.9 39.0 28.2 29.4 
RIDGE + Orthogonality 46.2 39.9 35.1 32.0 47.8 44.2 32.7 32.7 
OLS with L1 & L2 regularizers (E-NET) 42.1 36.8 29.3 30.1 42.2 38.2 28.5 28.5 
E-NET + Orthogonality 45.6 40.3 35.3 31.6 47.7 44.0 32.5 32.9 
View Large
Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy.
Accept