tensor model 0.002667466
vector model 0.002306156
linear model 0.002286263
model parameters 0.002204768
model complexity 0.002179703
space model 0.002095168
model regularization 0.002013382
sor model 0.0019799369999999998
model updates 0.0019787579999999997
model construction 0.001964629
model param 0.00196289
model regular 0.00196289
feature tensor 0.001947626
model 0.00173687
tensor models 0.001716317
training data 0.001656651
feature set 0.0016462529999999999
training algorithm 0.001618494
feature order 0.001600509
feature vector 0.001586316
large feature 0.001518058
feature weights 0.001515519
tensor order 0.0015140750000000001
order tensor 0.0015140750000000001
training set 0.001493434
linear tensor 0.0014799890000000001
feature weight 0.001467772
tensor representation 0.001397388
tensor weight 0.001381338
weight tensor 0.001381338
target feature 0.001379898
optimal tensor 0.0013671199999999999
such models 0.001346697
linear models 0.001335114
tensor structure 0.001314311
other words 0.001311142
training complexity 0.001307044
original feature 0.001297328
learning algorithm 0.0012944760000000001
tensor space 0.001288894
feature sets 0.001286478
tensor mapping 0.001279545
parameter number 0.00126542
features weights 0.001262505
true feature 0.001257902
good tensor 0.001255758
individual feature 0.0012542669999999999
inal feature 0.0012453269999999998
reranking feature 0.001242998
tensor decomposition 0.001239312
proper tensor 0.0011978940000000001
settings training 0.001197241
tensor entry 0.001194579
tensor representations 0.001188866
tensor mode 0.001188484
little training 0.0011846130000000002
tensor approximation 0.001175267
algorithm figure 0.001172026
tensor volume 0.001165934
tensor degrees 0.001161433
hth tensor 0.001155914
der tensor 0.001155914
core tensor 0.001155914
ing data 0.0011481590000000002
way features 0.001145633
space models 0.001144019
function value 0.001142561
training sets 0.001133659
algorithm performance 0.001129476
target features 0.0011268839999999999
labeled training 0.001112526
training environment 0.001110281
training com 0.0011054230000000001
traditional models 0.001096407
training samples 0.001093839
training epoch 0.001089808
data sets 0.001061888
data increases 0.001055016
models mira 0.001053779
set figure 0.001046966
approximate algorithm 0.00104093
labeled data 0.001040755
same time 0.0010400029999999998
function values 0.001034827
following order 0.001031426
large number 0.001027261
data sections 0.001023622
weight vector 0.001020028
feature 0.00101703
linguistic features 0.001007308
weight learning 9.90935E-4
objective function 9.68111E-4
other hand 9.527909999999999E-4
following problem 9.423470000000001E-4
possible values 9.405500000000001E-4
optimal value 9.34162E-4
tensor 9.30596E-4
vector space 9.27584E-4
evaluation set 9.0817E-4
space linear 9.07691E-4
