Skip to Main Content
Table 1: 
Hyperparameter values.
Hyperparametervalue
dw (English word embeddings) 100 
dw (other languages word embeddings) 300 
dc (character embeddings) 300 
dpos (POS embeddings) 16 
dl (lemma embeddings) 100 
dh (LSTM hidden states) 300 
dhidden (hidden layer representation) 200 
doutput (output label embeddings) 32 
dr (role representation) 128 
dl′ (output lemma representation) 128 
K (BiLSTM depth) 4 
J (BiLSTM depth) 2 
batch size 30 
input layer dropout rate 0.3 
hidden layer dropout rate 0.3 
learning rate 0.001 
auxiliary tasks loss weight α 0.5 
Hyperparametervalue
dw (English word embeddings) 100 
dw (other languages word embeddings) 300 
dc (character embeddings) 300 
dpos (POS embeddings) 16 
dl (lemma embeddings) 100 
dh (LSTM hidden states) 300 
dhidden (hidden layer representation) 200 
doutput (output label embeddings) 32 
dr (role representation) 128 
dl′ (output lemma representation) 128 
K (BiLSTM depth) 4 
J (BiLSTM depth) 2 
batch size 30 
input layer dropout rate 0.3 
hidden layer dropout rate 0.3 
learning rate 0.001 
auxiliary tasks loss weight α 0.5 
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