arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2406.16971stat.MLcs.LG

用于归一化流的灵活尾部

Flexible Tails for Normalizing Flows

School of Mathematics University of Bristol(数学学院 英国布里斯托尔大学)

Tennessee Hickling, Dennis Prangle

AI总结:

针对标准归一化流难以表示重尾分布的问题,提出使用高斯基分布加可生成重尾的最终变换层的尾部变换流(TTF),实验证明其在高维度或大尾重场景下性能优于现有重尾基分布方案。

摘要:

归一化流是一类灵活的概率分布,通过对简单基分布进行变换来表示。标准归一化流的一个局限是难以表示具有重尾的分布,这类分布在密度估计和变分推断的应用中均会出现。目前针对该问题的一种流行解决方案是使用重尾基分布。我们认为,由于在重尾输入下优化神经网络(例如归一化流)存在难度,这种方法可能导致性能不佳。我们提出了一种替代方案——“尾部变换流”(TTF),该方法使用高斯基分布,并搭配一个可产生重尾的最终变换层。实验结果表明,该方法优于现有方法,尤其是在目标分布维度较高或尾重较大时效果更显著。

英文摘要:

Normalizing flows are a flexible class of probability distributions, expressed as transformations of a simple base distribution. A limitation of standard normalizing flows is representing distributions with heavy tails, which arise in applications to both density estimation and variational inference. A popular current solution to this problem is to use a heavy tailed base distribution. We argue this can lead to poor performance due to the difficulty of optimising neural networks, such as normalizing flows, under heavy tailed input. We propose an alternative, "tail transform flow" (TTF), which uses a Gaussian base distribution and a final transformation layer which can produce heavy tails. Experimental results show this approach outperforms current methods, especially when the target distribution has large dimension or tail weight.

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。

↑