arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2107.08792cs.LGcs.AI
收藏

通过选择性数据匹配增强条件生成对抗网络训练的稳定性

Enhancing Stability in Training Conditional Generative Adversarial Networks via Selective Data Matching

Pusan National University(釜山大学);NHN Cloud(NHN云);Sogang University(西江大学)

Kyeongbo Kong, Kyunghun Kim, Suk-Ju Kang

AI总结:

本文提出选择性聚焦学习,通过按判别器条件项得分选择样本,仅对高分样本应用条件匹配,其余用联合匹配,在多个数据集上显著提升cGAN训练稳定性(FID最高改善35.18%)。

摘要:

条件生成对抗网络(cGANs)因其类别可控性和在复杂生成任务中的卓越质量而取得了显著成功。典型的cGANs通过将联合分布匹配问题分解为两个更简单的子问题来解决:边缘匹配和条件匹配。本文提出了一种简单但有效的训练方法,即选择性聚焦学习,该方法强制判别器和生成器快速学习每个类别的简单样本,同时保持多样性。我们的关键思想是对每个小批量中的数据有选择地应用条件匹配和联合匹配。具体来说,我们首先根据判别器输出的条件项(真实样本和生成样本)进行排序,选择得分最高的样本。然后,我们仅使用所选样本进行条件匹配,而使用其他样本进行联合匹配来优化模型。通过我们的玩具实验,我们发现由于判别器的内容感知优化,仅对某些样本应用条件匹配是最佳选择。我们在ImageNet(64x64和128x128)、CIFAR-10、CIFAR-100数据集以及混合高斯、噪声标签设置上进行了实验,以证明所提出的方法能够显著(在FID方面最高提升35.18%)改善所有指标,且进行了10次独立试验。代码可在https://github.com/pnu-cvsp/Enhancing-Stability-in-Training-Conditional-GAN-via-Selective-Data-Matching获取。

英文摘要:

Conditional generative adversarial networks (cGANs) have demonstrated remarkable success due to their class-wise controllability and superior quality for complex generation tasks. Typical cGANs solve the joint distribution matching problem by decomposing two easier sub-problems: marginal matching and conditional matching. In this paper, we proposes a simple but effective training methodology, selective focusing learning, which enforces the discriminator and generator to learn easy samples of each class rapidly while maintaining diversity. Our key idea is to selectively apply conditional and joint matching for the data in each mini-batch.Specifically, we first select the samples with the highest scores when sorted using the conditional term of the discriminator outputs (real and generated samples). Then we optimize the model using the selected samples with only conditional matching and the other samples with joint matching. From our toy experiments, we found that it is the best to apply only conditional matching to certain samples due to the content-aware optimization of the discriminator. We conducted experiments on ImageNet (64x64 and 128x128), CIFAR-10, CIFAR-100 datasets, and Mixture of Gaussian, noisy label settings to demonstrate that the proposed method can substantially (up to 35.18% in terms of FID) improve all indicators with 10 independent trials. Code is available at https://github.com/pnu-cvsp/Enhancing-Stability-in-Training-Conditional-GAN-via-Selective-Data-Matching.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑