Mining GOLD Samples for Conditional GANs
This work addresses the challenge of enhancing training, inference, and data selection for cGANs, which are used in image generation tasks, but it is incremental as it builds on existing cGAN frameworks.
The paper tackles the problem of improving conditional generative adversarial networks (cGANs) by introducing a measure called GOLD to assess the discrepancy between data and model distributions, and it shows that applications like example re-weighting, rejection sampling, and active learning outperform baselines on various image datasets.
Conditional generative adversarial networks (cGANs) have gained a considerable attention in recent years due to its class-wise controllability and superior quality for complex generation tasks. We introduce a simple yet effective approach to improving cGANs by measuring the discrepancy between the data distribution and the model distribution on given samples. The proposed measure, coined the gap of log-densities (GOLD), provides an effective self-diagnosis for cGANs while being efficienty computed from the discriminator. We propose three applications of the GOLD: example re-weighting, rejection sampling, and active learning, which improve the training, inference, and data selection of cGANs, respectively. Our experimental results demonstrate that the proposed methods outperform corresponding baselines for all three applications on different image datasets.