feature selection 0.003504943
feature set 0.003441668
gradient feature 0.003366675
feature gradient 0.003366675
feature space 0.003283573
simple feature 0.0031935080000000003
feature sets 0.003185073
feature constraints 0.003172391
feature templates 0.003171928
efficient feature 0.0031704050000000003
large feature 0.003166527
feature count 0.0031599920000000004
incremental feature 0.003147891
feature functions 0.003144384
full feature 0.003136101
feature tests 0.003123725
few feature 0.003116069
feature expectations 0.0031034400000000003
feature sparsity 0.003099727
feature cutoff 0.003090695
feature induction 0.003087606
feature inclusion 0.003087451
feature counts 0.003083473
feature testing 0.0030830740000000003
mental feature 0.003080897
motivated feature 0.00308003
standardized feature 0.003079067
tal feature 0.003076407
feature design 0.003075312
feature construction 0.003073845
feature tem 0.0030732980000000003
feature gradients 0.003073204
feature detection 0.003072659
dient feature 0.003072243
feature 0.00286022
active features 0.002727258
features sets 0.002718453
large features 0.002699907
linguistic features 0.002697166
iterations features 0.002673862
features times 0.002650569
motivated features 0.00261341
irrelevant features 0.002610746
redundant features 0.002608077
dundant features 0.0026039550000000002
former features 0.002603022
features forests 0.002603022
features 0.0023936
linear model 0.001880115
model distribution 0.001815419
model parameters 0.0017996050000000001
optimal model 0.001683108
label label 0.001671522
training data 0.001447994
model 0.00143472
data set 0.001325093
training set 0.001285797
test data 0.001215685
label constituent 0.00112153
constituent label 0.00112153
input data 0.001097538
linear models 0.0010820130000000002
label child 0.0010673129999999999
child label 0.0010673129999999999
descendant label 0.001056437
label descendant 0.001056437
parent label 0.001054306
test set 0.001053488
label ancestor 0.001049859
ancestor label 0.001049859
embedded label 0.001045598
probability function 0.001040665
same time 0.001040016
likelihood function 0.001035377
objective function 0.001015902
selection step 9.99275E-4
ture selection 9.9602E-4
data points 9.95552E-4
value value 9.90016E-4
regularization method 9.84141E-4
convex function 9.83504E-4
optimization algorithm 9.83228E-4
gradient test 9.78495E-4
final training 9.765169999999999E-4
experimental results 9.75852E-4
training examples 9.66003E-4
data program 9.646E-4
heldout data 9.59752E-4
lasso algorithm 9.546890000000001E-4
massive data 9.53189E-4
original function 9.50154E-4
parameter values 9.44252E-4
dual function 9.43286E-4
parameter weights 9.38187E-4
different size 9.37137E-4
potential function 9.33639E-4
tive function 9.33138E-4
parse selection 9.33014E-4
sparse models 9.324930000000001E-4
tial function 9.16371E-4
