Pairing Regularization for Mitigating Many-to-One Collapse in GANs
This addresses a specific, understudied failure mode in GAN training for generative modeling, offering a complementary technique to existing stabilization methods.
The paper tackles the problem of intra-mode collapse in GANs, where many latent variables map to similar outputs, by proposing a pairing regularizer that enforces local consistency between latents and generated samples, leading to improved coverage, recall, and precision in experiments on toy distributions and real-image benchmarks.
Mode collapse remains a fundamental challenge in training generative adversarial networks (GANs). While existing works have primarily focused on inter-mode collapse, such as mode dropping, intra-mode collapse-where many latent variables map to the same or highly similar outputs-has received significantly less attention. In this work, we propose a pairing regularizer jointly optimized with the generator to mitigate the many-to-one collapse by enforcing local consistency between latent variables and generated samples. We show that the effect of pairing regularization depends on the dominant failure mode of training. In collapse-prone regimes with limited exploration, pairing encourages structured local exploration, leading to improved coverage and higher recall. In contrast, under stabilized training with sufficient exploration, pairing refines the generator's induced data density by discouraging redundant mappings, thereby improving precision without sacrificing recall. Extensive experiments on both toy distributions and real-image benchmarks demonstrate that the proposed regularizer effectively complements existing stabilization techniques by directly addressing intra-mode collapse.