Good Semi-supervised Learning that Requires a Bad GAN

NeurIPS 2017 Zihang DaiZhilin YangFan YangWilliam W. CohenRuslan Salakhutdinov

Semi-supervised learning methods based on generative adversarial networks (GANs) obtained strong empirical results, but it is not clear 1) how the discriminator benefits from joint training with a generator, and 2) why good semi-supervised classification performance and a good generator cannot be obtained at the same time. Theoretically, we show that given the discriminator objective, good semisupervised learning indeed requires a bad generator, and propose the definition of a preferred generator... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Semi-Supervised Image Classification CIFAR-10, 4000 Labels Bad GAN Accuracy 85.59 # 15