arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.09923stat.MLcs.LGmath.OC
收藏

Tessellated Wasserstein自动编码器

Tessellated Wasserstein Auto-Encoders

Kuo Gai, Shihua Zhang

摘要:

非对抗生成模型,如变分自动编码器(VAE),Wasserstein自动编码器与最大平均差异(WAE-MMD),切片Wasserstein自动编码器(SWAE)是相对容易训练,并具有较少的模式崩溃相比,Wasserstein自动编码器与生成对抗网络(WAE-GAN)。然而,它们在近似潜在空间中的目标分布方面不是很准确,因为它们没有一个用于检测真假之间的微小差异的滤波器。为此,我们开发了一种称为Tessellated Wasserstein Auto-encoders(TWAE)的新型非对抗性框架,通过质心Voronoi细分(CVT)技术将目标分布的支持细分为给定数量的区域,并根据细分设计批量数据,而不是随机洗牌,以准确计算差异。理论上,我们证明了当镶嵌的样本数$n$和区域数$m$分别以$\mathcal{O}(\frac{1}{\sqrt{n}})$和$\mathcal{O}(\frac{1}{\sqrt{m}})$的速率变大时,对差异的估计误差减小。给定固定的$n$和$m$,使测量误差的上界最小化的必要条件是镶嵌是由CVT确定的镶嵌。TWAE对不同的非对抗性指标非常灵活,并且与VAE,WAE-MMD,SWAE相比,可以大大提高其在Fréchet起始距离(FID)方面的生成性能。此外,数值结果确实表明TWAE与对抗模型WAE-GAN竞争,证明了其强大的生成能力。

英文摘要:

Non-adversarial generative models such as variational auto-encoder (VAE), Wasserstein auto-encoders with maximum mean discrepancy (WAE-MMD), sliced-Wasserstein auto-encoder (SWAE) are relatively easy to train and have less mode collapse compared to Wasserstein auto-encoder with generative adversarial network (WAE-GAN). However, they are not very accurate in approximating the target distribution in the latent space because they don't have a discriminator to detect the minor difference between real and fake. To this end, we develop a novel non-adversarial framework called Tessellated Wasserstein Auto-encoders (TWAE) to tessellate the support of the target distribution into a given number of regions by the centroidal Voronoi tessellation (CVT) technique and design batches of data according to the tessellation instead of random shuffling for accurate computation of discrepancy. Theoretically, we demonstrate that the error of estimate to the discrepancy decreases when the numbers of samples $n$ and regions $m$ of the tessellation become larger with rates of $\mathcal{O}(\frac{1}{\sqrt{n}})$ and $\mathcal{O}(\frac{1}{\sqrt{m}})$, respectively. Given fixed $n$ and $m$, a necessary condition for the upper bound of measurement error to be minimized is that the tessellation is the one determined by CVT. TWAE is very flexible to different non-adversarial metrics and can substantially enhance their generative performance in terms of Fréchet inception distance (FID) compared to VAE, WAE-MMD, SWAE. Moreover, numerical results indeed demonstrate that TWAE is competitive to the adversarial model WAE-GAN, demonstrating its powerful generative ability.

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。

↑