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 1906.04516stat.MLcs.LG
收藏

切片-瓦瑟斯坦距离下生成模型学习的渐近保证

Asymptotic Guarantees for Learning Generative Models with the Sliced-Wasserstein Distance

Télécom Paris(巴黎电信学院);Institut Polytechnique de Paris(巴黎理工学院);ENS Cachan(卡尚高等师范学校);CNRS(法国国家科学研究中心);Université Paris-Saclay(巴黎萨克雷大学);University of Oxford(牛津大学)

Kimia Nadjahi, Alain Durmus, Umut Şimşekli, Roland Badeau

摘要:

最小期望距离估计(MEDE)算法已被广泛用于具有难处理的似然函数的概率模型,并且由于其在隐式生成建模(例如Wasserstein生成对抗网络,Wasserstein自编码器)中的使用,它们已变得越来越流行。从计算最优传输中产生的切片-瓦瑟斯坦(SW)距离由于其简单性和计算优势已成为MEDE中的流行选择。虽然有几项研究已经报道了经验的成功与SW生成建模,这种估计的理论属性尚未建立。在这项研究中,我们调查的渐近性质的估计,通过最小化SW。我们首先表明,在SW收敛意味着弱收敛的概率测度在一般Wasserstein空间。然后,我们证明了通过最小化SW(以及SW的近似版本)获得的估计量是渐近一致的。最后我们证明了一个中心极限定理,它刻画了估计量的渐近分布,并建立了$\sqrt{n}$的收敛速度,其中$n$表示观测数据点的个数.我们说明了我们的理论在合成数据和神经网络上的有效性。

英文摘要:

Minimum expected distance estimation (MEDE) algorithms have been widely used for probabilistic models with intractable likelihood functions and they have become increasingly popular due to their use in implicit generative modeling (e.g. Wasserstein generative adversarial networks, Wasserstein autoencoders). Emerging from computational optimal transport, the Sliced-Wasserstein (SW) distance has become a popular choice in MEDE thanks to its simplicity and computational benefits. While several studies have reported empirical success on generative modeling with SW, the theoretical properties of such estimators have not yet been established. In this study, we investigate the asymptotic properties of estimators that are obtained by minimizing SW. We first show that convergence in SW implies weak convergence of probability measures in general Wasserstein spaces. Then we show that estimators obtained by minimizing SW (and also an approximate version of SW) are asymptotically consistent. We finally prove a central limit theorem, which characterizes the asymptotic distribution of the estimators and establish a convergence rate of $\sqrt{n}$, where $n$ denotes the number of observed data points. We illustrate the validity of our theory on both synthetic data and neural networks.

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。

↑