A regularized self-training framework that softens pseudo-labels or network outputs improves unsupervised domain adaptation over the class-balanced self-training baseline.
Progressive feature alignment for unsupervised do- main adaptation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Confidence Regularized Self-Training
A regularized self-training framework that softens pseudo-labels or network outputs improves unsupervised domain adaptation over the class-balanced self-training baseline.