arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.13913stat.MLcs.LG
收藏

同时流形学习和密度估计的流

Flows for simultaneous manifold learning and density estimation

New York University(纽约大学)

Johann Brehmer, Kyle Cranmer

摘要:

我们引入流形学习流(M流),一类新的生成模型,同时学习数据流形以及该流形上的易处理的概率密度。结合规范化流、GANs、自动编码器和基于能量的模型的各个方面,它们有可能更忠实地表示具有流形结构的数据集,并提供降维、去噪和分布外检测的处理。我们认为,为什么这样的模型不应该单独训练的极大似然,并提出了一个新的训练算法,分离流形和密度更新。在一系列实验中,我们展示了M-flow如何学习数据流形,并允许比环境数据空间中的标准流更好的推理。

英文摘要:

We introduce manifold-learning flows (M-flows), a new class of generative models that simultaneously learn the data manifold as well as a tractable probability density on that manifold. Combining aspects of normalizing flows, GANs, autoencoders, and energy-based models, they have the potential to represent datasets with a manifold structure more faithfully and provide handles on dimensionality reduction, denoising, and out-of-distribution detection. We argue why such models should not be trained by maximum likelihood alone and present a new training algorithm that separates manifold and density updates. In a range of experiments we demonstrate how M-flows learn the data manifold and allow for better inference than standard flows in the ambient data space.

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。

↑