PConv: Simple yet Effective Convolutional Layer for Generative Adversarial Network

TL;DR

This paper introduces PConv, a simple yet effective convolutional layer for GANs that improves performance and reduces discriminator memorization by perturbing input features during training.

Contribution

The paper proposes a novel perturbed convolution layer that enhances GAN training stability and generalization by incorporating random feature disturbances.

Findings

PConv improves GAN performance across multiple datasets.

PConv reduces discriminator memorization in training.

PConv enhances robustness of learned features.

Abstract

This paper presents a novel convolutional layer, called perturbed convolution (PConv), which focuses on achieving two goals simultaneously: improving the generative adversarial network (GAN) performance and alleviating the memorization problem in which the discriminator memorizes all images from a given dataset as training progresses. In PConv, perturbed features are generated by randomly disturbing an input tensor before performing the convolution operation. This approach is simple but surprisingly effective. First, to produce a similar output even with the perturbed tensor, each layer in the discriminator should learn robust features having a small local Lipschitz value. Second, since the input tensor is randomly perturbed during the training procedure like the dropout in neural networks, the memorization problem could be alleviated. To show the generalization ability of the proposed method, we conducted extensive experiments with various loss functions and datasets including CIFAR-10, CelebA, CelebA-HQ, LSUN, and tiny-ImageNet. The quantitative evaluations demonstrate that PConv effectively boosts the performance of GAN and conditional GAN in terms of Frechet inception distance (FID).