arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.03575stat.MLcs.LG
收藏

有限样本复杂度下基于流的生成模型学习问题分析

Analysis of learning a flow-based generative model from limited sample complexity

École Polytechnique Fédérale de Lausanne(洛桑联邦理工学院)

Hugo Cui, Florent Krzakala, Eric Vanden-Eijnden, Lenka Zdeborová

AI总结:

该研究针对两层自编码器参数化的流生成模型在高维高斯混合采样任务中的有限样本学习问题,给出速度场与生成流的精准闭式分析,证明均值距离以$Θ_n(1/n)$速率衰减且为贝叶斯最优。

摘要:

我们研究训练以两层自编码器为参数化形式的基于流的生成模型,用于从高维高斯混合分布中采样的问题,并对该问题给出了精准的端到端分析。首先,当使用目标分布的有限数量$n$个样本训练浅层去噪自编码器作为参数化结构时,我们对学习到的速度场给出了紧致的闭式刻画。基于该分析,我们对将基础高斯密度前推为目标密度近似的对应生成流给出了精准描述。具体而言,我们推导出生成混合分布的均值与目标混合分布的均值之间距离的闭式公式,并证明该距离以$Θ_n(\ rac{1}{n})$的速率衰减。最后,我们证明该速率实际上是贝叶斯最优的。

英文摘要:

We study the problem of training a flow-based generative model, parametrized by a two-layer autoencoder, to sample from a high-dimensional Gaussian mixture. We provide a sharp end-to-end analysis of the problem. First, we provide a tight closed-form characterization of the learnt velocity field, when parametrized by a shallow denoising auto-encoder trained on a finite number $n$ of samples from the target distribution. Building on this analysis, we provide a sharp description of the corresponding generative flow, which pushes the base Gaussian density forward to an approximation of the target density. In particular, we provide closed-form formulae for the distance between the mean of the generated mixture and the mean of the target mixture, which we show decays as $Θ_n(\frac{1}{n})$. Finally, this rate is shown to be in fact Bayes-optimal.

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。

↑