arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.14151stat.MEcs.AIcs.LGstat.ML
收藏

现代非线性函数对函数回归

Modern Non-Linear Function-on-Function Regression

Pennsylvania State University(宾夕法尼亚州立大学)

Aniruddha Rajendra Rao, Matthew Reimherr

AI总结:

本研究针对函数型数据提出基于神经网络的非线性函数对函数回归框架,含连续隐藏层与两种拟合策略,通过函数梯度和正则化拟合,经模拟与真实数据验证了方法的效力与灵活性。

摘要:

我们针对函数型数据,利用神经网络提出了一类新的非线性函数对函数回归模型。我们提出了一种采用由连续神经元构成的隐藏层(称为连续隐藏层)的框架用于函数型响应建模,并给出两种模型拟合策略:Functional Direct Neural Network(FDNN,函数型直接神经网络)和Functional Basis Neural Network(FBNN,函数型基神经网络)。二者均被专门设计用于利用函数型数据的固有结构,捕捉函数型预测变量与函数型响应之间存在的复杂关系。我们通过推导函数型梯度来拟合这些模型,并应用正则化技术以获得更简约的结果。我们通过大量模拟研究以及真实数据示例,证明了所提方法在处理复杂函数型模型方面的效力与灵活性。

英文摘要:

We introduce a new class of non-linear function-on-function regression models for functional data using neural networks. We propose a framework using a hidden layer consisting of continuous neurons, called a continuous hidden layer, for functional response modeling and give two model fitting strategies, Functional Direct Neural Network (FDNN) and Functional Basis Neural Network (FBNN). Both are designed explicitly to exploit the structure inherent in functional data and capture the complex relations existing between the functional predictors and the functional response. We fit these models by deriving functional gradients and implement regularization techniques for more parsimonious results. We demonstrate the power and flexibility of our proposed method in handling complex functional models through extensive simulation studies as well as real data examples.

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。

↑