logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
delete
arrow
返回

Deep learning with missing data

delete2026-07-29
delete0
delete
OA
AI
T
Tianyi Ma
T
Tengyao Wang
R
Richard J. Samworth *
DOI：10.1093/jrsssb/qkag114delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
在多元非参数回归且协变量缺失的背景下，我们提出了**模式嵌入神经网络**（PENNs），它可以与任何现有的插补技术结合使用。除了在插补数据上训练的神经网络外，PENNs还将观测指标向量通过第二个神经网络传递，以提供紧凑表示。然后，这些输出在第三个神经网络中结合，以产生最终预测。我们的主要理论结果利用了一个假设，即观测模式可以被划分为若干单元，在这些单元上贝叶斯回归函数的行为相似，且属于复合Hölder类。它提供了一个有限样本的风险上界，适用于任意缺失机制，并结合一个补充的极小极大下界，表明我们的PENN估计量在典型情况下能达到极小极大收敛率，如同分区单元预先已知一样，最多相差一个样本量对数多项式因子。在模拟、半合成和真实数据上的数值实验证实，PENN估计量持续性地改善，常常显著优于未嵌入模式的常规神经网络。用于复现我们实验的代码以及如何应用我们方法的教程已公开可用。

期刊

J
Journal of the Royal Statistical Society Series B: Statistical Methodology
IF:
0
论文数:
111
被引数:
0

机构

L
London School of Economics and Political Science
学者数:
23
论文数: 20
被引数: 0
U
University of Cambridge
学者数:
986
论文数: 474
被引数: 0
引用论文

引用论文

High-dimensional principal component analysis with heterogeneous missingness
errJOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
IF3.6
err2022-11-20
err12
errOAAI
errZhu, Ziwei; Wang, Tengyao; Samworth, Richard J.
err分享
err收藏
The Generalization Error of Random Features Regression: Precise Asymptotics and the Double Descent Curve随机特征回归的泛化误差: 精确渐近和双下降曲线
errCOMMUNICATIONS ON PURE AND APPLIED MATHEMATICS
IF2.7
err2021-06-06
err168
errOAAI
errMei, Song; Montanari, Andrea
err分享
err收藏
Sparse spectral estimation with missing and corrupted measurements
errStat
IF0
err2019-06-26
err0
errOAAI
errAndreas Elsener; Sara van de Geer
err分享
err收藏
A Selective Overview of Deep Learning
errSTATISTICAL SCIENCE
IF3.4
err2021-05-01
err72
errOAAI
errFan, Jianqing; Ma, Cong; Zhong, Yiqiao
err分享
err收藏
High-dimensional changepoint estimation with heterogeneous missingness具有异质错失度的高维变化点估计
errJOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
IF3.6
err2022-07-11
err7
errOAAI
errFollain, Bertille; Wang, Tengyao; Samworth, Richard J.
err分享
err收藏
mice: Multivariate Imputation by Chained Equations inR 小鼠 : 通过 R 中的链式方程进行多元插补
errJournal of Statistical Software
IF0
err2011-01-01
err0
errOAAI
errStef van Buuren; Karin Groothuis-Oudshoorn
err分享
err收藏
On the consistency of supervised learning with missing values
errStatistical Papers
IF0
err2024-09-12
err0
errOAAI
errJulie Josse; Jacob M. Chen; Nicolas Prost; Gaël Varoquaux; Erwan Scornet
err分享
err收藏
A supervised deep learning method for nonparametric density estimation
errElectronic Journal of Statistics
IF0
err2024-01-01
err0
PREAI
errBos,Thijs; Schmidt-Hieber,Johannes
err分享
err收藏
OPTIMAL NONPARAMETRIC TESTING OF MISSING COMPLETELY AT RANDOM AND ITS CONNECTIONS TO COMPATIBILITY
errANNALS OF STATISTICS
IF3.7
err2023-10-01
err4
errOAAI
errBerrett, Thomas b.; Samworth, Richard j.
err分享
err收藏
Highly accurate protein structure prediction with AlphaFold基于AlphaFold的高精度蛋白质结构预测
errNature
IF0
err2021-07-15
err0
errOAAI
errJohn Jumper; Richard Evans; Alexander Pritzel; Tim Green; Michael Figurnov; Olaf Ronneberger; Kathryn Tunyasuvunakool; Russ Bates; Augustin Žídek; Anna Potapenko; Alex Bridgland; Clemens Meyer; Simon A. A. Kohl; Andrew J. Ballard; Andrew Cowie; Bernardino Romera-Paredes; Stanislav Nikolov; Rishub Jain; Jonas Adler; Trevor Back; Stig Petersen; David Reiman; Ellen Clancy; Michal Zielinski; Martin Steinegger; Michalina Pacholska; Tamas Berghammer; Sebastian Bodenstein; David Silver; Oriol Vinyals; Andrew W. Senior; Koray Kavukcuoglu; Pushmeet Kohli; Demis Hassabis
err分享
err收藏
学者 查看更多内容