Skip to Main Content
Table 8 
Analysis of refinement methods applied to the same initial mappings of our adversarial autoencoder on the Conneau data set for Ar, Ms, and He languages.
 En-ArEn-MsEn-He
Artetxe refinement 
Robust self-learning 35.6 51.7 52.2 50.5 43.7 56.3 
Robust self-learning + Symmetric re-weighting 38.3 54.1 54.0 54.4 44.9 58.1 
  
Our refinement 
Procrustes solution 33.8 49.9 49.5 48.6 41.1 56.8 
Symmetric re-weighting 36.1 54.0 54.6 55.0 45.8 57.1 
Procrustes solution + Symmetric re-weighting 38.6 55.7 54.8 55.2 46.1 58.6 
 En-ArEn-MsEn-He
Artetxe refinement 
Robust self-learning 35.6 51.7 52.2 50.5 43.7 56.3 
Robust self-learning + Symmetric re-weighting 38.3 54.1 54.0 54.4 44.9 58.1 
  
Our refinement 
Procrustes solution 33.8 49.9 49.5 48.6 41.1 56.8 
Symmetric re-weighting 36.1 54.0 54.6 55.0 45.8 57.1 
Procrustes solution + Symmetric re-weighting 38.6 55.7 54.8 55.2 46.1 58.6 
Close Modal

or Create an Account

Close Modal
Close Modal