MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators
This addresses a chronic issue in generative modeling for researchers and practitioners, though it appears incremental as it builds on existing GAN architectures with a multi-discriminator approach.
The paper tackles the mode collapse problem in GANs by proposing a framework with multiple specialized discriminators that collaborate to better represent the real data distribution, achieving improved performance on standard datasets as measured by multiple evaluation metrics.
We propose a framework of generative adversarial networks with multiple discriminators, which collaborate to represent a real dataset more effectively. Our approach facilitates learning a generator consistent with the underlying data distribution based on real images and thus mitigates the chronic mode collapse problem. From the inspiration of multiple choice learning, we guide each discriminator to have expertise in a subset of the entire data and allow the generator to find reasonable correspondences between the latent and real data spaces automatically without extra supervision for training examples. Despite the use of multiple discriminators, the backbone networks are shared across the discriminators and the increase in training cost is marginal. We demonstrate the effectiveness of our algorithm using multiple evaluation metrics in the standard datasets for diverse tasks.