feature selection 0.002930308
feature weights 0.002903717
feature space 0.0028287349999999998
feature weight 0.002824471
additional feature 0.002760892
feature templates 0.0027576379999999998
entire feature 0.002729904
feature template 0.002710073
group features 0.0026923909999999997
feature engineering 0.002668285
feature levels 0.0026501059999999997
contextual feature 0.002649171
feature tem 0.002641381
manual feature 0.0026341579999999997
other features 0.002580101
many features 0.002424825
feature 0.00242386
input features 0.002348732
few features 0.002310004
unigram features 0.00225831
individual features 0.0022524389999999997
cal features 0.002224439
candidate features 0.002221191
lexical features 0.002210275
speech features 0.002203475
irrelevant features 0.0022033739999999997
finer features 0.00219721
features 0.0019779
model size 0.0015542560000000002
model parameters 0.0015375500000000002
model sparsity 0.00148441
probabilistic model 0.00145809
current model 0.001447876
model sizes 0.001434897
bayes model 0.0014322710000000002
model design 0.001389849
model interpretability 0.0013811910000000001
model designer 0.001379591
learning algorithm 0.001345962
loss function 0.0013085129999999999
previous word 0.001243109
model 0.00116871
gradient algorithm 0.001156537
learning problem 0.001155087
linear models 0.0011325390000000001
several group 0.001124254
kernel learning 0.001115139
cost function 0.001111581
dependency parsing 0.001105432
group size 0.001100037
word shape 0.001084459
word shapes 0.001081407
structured groups 0.001077681
objective function 0.001062225
lasso group 0.001057304
group lasso 0.001057304
same size 0.001039226
label bigrams 0.0010311370000000001
label trigrams 0.001023848
entire group 0.001020535
mira group 0.001016213
training pair 0.001013676
training sample 0.001012939
function proxω 0.001012853
single group 0.001010043
structure learning 0.001007254
learning problems 0.001002127
similar algorithm 9.96949E-4
group choices 9.79192E-4
graphical models 9.67253E-4
ative models 9.59708E-4
gradient vector 9.536600000000001E-4
online algorithm 9.52247E-4
unseen data 9.487759999999999E-4
different regularization 9.34888E-4
tain models 9.32878E-4
input set 9.32803E-4
mira algorithm 9.319700000000001E-4
pact models 9.31419E-4
supervised learning 9.29379E-4
criminative learning 9.27408E-4
identical learning 9.26467E-4
multitask learning 9.26467E-4
group indices 9.26402E-4
turkish group 9.258370000000001E-4
group decompo 9.258370000000001E-4
singleton group 9.258370000000001E-4
task datasets 9.23636E-4
structured case 9.168029999999999E-4
several loss 9.166560000000001E-4
same idea 9.146530000000001E-4
output labels 9.130189999999999E-4
active set 9.10619E-4
gradient methods 9.09562E-4
regularization term 9.04906E-4
dependency parsers 9.03808E-4
few groups 8.939099999999999E-4
optimization problem 8.91134E-4
same plate 8.84391E-4
only work 8.81884E-4
