Domain Partitioning Network
This addresses mode collapse in GANs, a common issue in generative modeling, though it appears incremental by building on existing multi-discriminator approaches.
The paper tackles mode collapse in generative adversarial networks by introducing the Domain Partitioning Network (DoPaNet), which uses multiple discriminators and a classifier to partition the target distribution, resulting in improved coverage and control over generated samples compared to competing methods.
Standard adversarial training involves two agents, namely a generator and a discriminator, playing a mini-max game. However, even if the players converge to an equilibrium, the generator may only recover a part of the target data distribution, in a situation commonly referred to as mode collapse. In this work, we present the Domain Partitioning Network (DoPaNet), a new approach to deal with mode collapse in generative adversarial learning. We employ multiple discriminators, each encouraging the generator to cover a different part of the target distribution. To ensure these parts do not overlap and collapse into the same mode, we add a classifier as a third agent in the game. The classifier decides which discriminator the generator is trained against for each sample. Through experiments on toy examples and real images, we show the merits of DoPaNet in covering the real distribution and its superiority with respect to the competing methods. Besides, we also show that we can control the modes from which samples are generated using DoPaNet.