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 2007.15255cs.CVcs.LGstat.ML
收藏

GANs的实例选择

Instance Selection for GANs

University of Guelph(圭尔夫大学);Vector Institute(向量研究所);Facebook AI Research(Facebook人工智能研究院)

Terrance DeVries, Michal Drozdzal, Graham W. Taylor

摘要:

生成对抗网络(GANs)的最新进展使其被广泛采用,以生成高质量的合成图像。虽然能够生成照片般逼真的图像,但这些模型通常会产生不切实际的样本,这些样本落在数据流形之外。最近提出的几种技术试图避免虚假样本,或者通过在生成后拒绝它们,或者通过截断模型的潜在空间。虽然有效,但这些方法效率低下,因为大部分训练时间和模型容量都用于最终未使用的样本。在这项工作中,我们提出了一种新的方法来提高样本质量:在模型训练之前通过实例选择来改变训练数据集。通过在训练前优化经验数据分布,我们将模型容量重定向到高密度区域,最终提高了样本保真度,降低了模型容量要求,并显著缩短了训练时间。代码可在https://github.com/uoguelph-mlrg/instance_selection_for_gans上获得。

英文摘要:

Recent advances in Generative Adversarial Networks (GANs) have led to their widespread adoption for the purposes of generating high quality synthetic imagery. While capable of generating photo-realistic images, these models often produce unrealistic samples which fall outside of the data manifold. Several recently proposed techniques attempt to avoid spurious samples, either by rejecting them after generation, or by truncating the model's latent space. While effective, these methods are inefficient, as a large fraction of training time and model capacity are dedicated towards samples that will ultimately go unused. In this work we propose a novel approach to improve sample quality: altering the training dataset via instance selection before model training has taken place. By refining the empirical data distribution before training, we redirect model capacity towards high-density regions, which ultimately improves sample fidelity, lowers model capacity requirements, and significantly reduces training time. Code is available at https://github.com/uoguelph-mlrg/instance_selection_for_gans.

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。

↑