search

blackfeather-wang / ISDA-for-Deep-Networks

An efficient implicit semantic augmentation method, complementary to existing non-semantic techniques.
582 stars 93 forks source link

confused about where the augmentation is done #34

Closed sungmin9939 closed 1 year ago

sungmin9939 commented 1 year ago

Hi

I read your paper. it's a very wonderful work

what i am confusing is related to #33

in #33, you said that because last fc layer(classifier) is implemented outside of whole model so y is embedding feature not logit

then why you update covariance using features variable not the y variable before entering isda_aug method?

what i understood by reading your paper is that the proposed augmentation is done on the embedding feature ai with covariance matrix yi which is constructed by calculating element wise variance of features whose classes are yi. am i right?

according to my understanding, covariance should be updated by using y variable

Thanks a lot!

sungmin9939 commented 1 year ago

sorry never mind

I figured it out by carefully reading equation 7 again in the paper