LLP-GAN trains a GAN discriminator to be the instance classifier using only bag-level label proportions, and its optimal discriminator is shown to be a bag-proportion mixture.
Ardehaly and Aron Culotta
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Learning from Label Proportions with Generative Adversarial Networks
LLP-GAN trains a GAN discriminator to be the instance classifier using only bag-level label proportions, and its optimal discriminator is shown to be a bag-proportion mixture.