arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.03487stat.MLcs.LG
收藏

基于神经网络的分布回归学习理论

Learning Theory of Distribution Regression with Neural Networks

KU Leuven(鲁汶大学);Hong Kong Baptist University(香港浸会大学);University of Sydney(悉尼大学)

Zhongjie Shi, Zhan Yu, Ding-Xuan Zhou

AI总结:

本文针对分布回归中输入为概率测度导致传统神经网络无法直接使用的问题,提出一种新颖的全连接神经网络框架,建立Borel概率测度空间上泛函的逼近理论,并通过两阶段误差分解技术推导出几乎最优的学习率。

摘要:

本文旨在通过全连接神经网络(FNN)建立分布回归的逼近理论与学习理论。与经典回归方法不同,分布回归的输入变量是概率测度。因此,我们通常需要进行第二阶段采样过程来近似分布的实际信息。另一方面,经典神经网络结构要求输入变量为向量。当输入样本是概率分布时,传统的深度神经网络方法无法直接使用,分布回归由此产生困难。一个针对分布输入的良好定义的神经网络结构是迫切需要的。目前尚无关于神经网络实现分布回归的数学模型和理论分析。为克服技术困难并解决这一问题,我们建立了一个新颖的全连接神经网络框架,以实现定义在Borel概率测度空间上泛函的逼近理论。此外,基于所建立的泛函逼近结果,在由具有分布输入的新颖FNN结构所诱导的假设空间中,通过一种新颖的两阶段误差分解技术,推导出了所提出的分布回归模型几乎最优的学习率(最多相差对数项)。

英文摘要:

In this paper, we aim at establishing an approximation theory and a learning theory of distribution regression via a fully connected neural network (FNN). In contrast to the classical regression methods, the input variables of distribution regression are probability measures. Then we often need to perform a second-stage sampling process to approximate the actual information of the distribution. On the other hand, the classical neural network structure requires the input variable to be a vector. When the input samples are probability distributions, the traditional deep neural network method cannot be directly used and the difficulty arises for distribution regression. A well-defined neural network structure for distribution inputs is intensively desirable. There is no mathematical model and theoretical analysis on neural network realization of distribution regression. To overcome technical difficulties and address this issue, we establish a novel fully connected neural network framework to realize an approximation theory of functionals defined on the space of Borel probability measures. Furthermore, based on the established functional approximation results, in the hypothesis space induced by the novel FNN structure with distribution inputs, almost optimal learning rates for the proposed distribution regression model up to logarithmic terms are derived via a novel two-stage error decomposition technique.

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。

↑