Timezone: »

Improving GANs with A Dynamic Discriminator
Ceyuan Yang · Yujun Shen · Yinghao Xu · Deli Zhao · Bo Dai · Bolei Zhou

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #128

Discriminator plays a vital role in training generative adversarial networks (GANs) via distinguishing real and synthesized samples. While the real data distribution remains the same, the synthesis distribution keeps varying because of the evolving generator, and thus effects a corresponding change of the bi-classification task assigned to the discriminator. We argue that a discriminator with an on-the-fly adjustment on its capacity can better accommodate such a time-varying task. A comprehensive empirical study confirms that the proposed training strategy, termed as DynamicD, improves the synthesis performance without incurring any additional computation cost or training objectives. Two capacity adjusting schemes are developed for training GANs under different data regimes: i) given a sufficient amount of training data, the discriminator benefits from a progressively increased learning capacity, and ii) when the training data is limited, gradually decreasing the layer width mitigates the over-fitting issue of the discriminator. Experiments on both 2D and 3D-aware image synthesis tasks conducted on a range of datasets substantiate the generalizability of our DynamicD as well as its substantial improvement over the baselines. Furthermore, DynamicD is synergistic to other discriminator-improving approaches (including data augmentation, regularizers, and pre-training), and brings continuous performance gain when combined with them for learning GANs. Code will be made publicly available.

Author Information

Ceyuan Yang (The Chinese University of Hong Kong)
Yujun Shen (Ant Research)
Yinghao Xu (Chinese University of Hong Kong)
Deli Zhao (Xiaomi AI Lab)
Bo Dai (Shanghai AI Lab)
Bolei Zhou (UCLA)
Bolei Zhou

Assistant professor at UCLA's computer science department

More from the Same Authors