融合套索加法模型
Fused Lasso Additive Model.
作者信息
Petersen Ashley, Witten Daniela, Simon Noah
机构信息
Department of Biostatistics, University of Washington, Seattle WA 98195.
出版信息
J Comput Graph Stat. 2016;25(4):1005-1025. doi: 10.1080/10618600.2015.1073155. Epub 2016 Nov 10.
Abstract
We consider the problem of predicting an outcome variable using covariates that are measured on independent observations, in a setting in which additive, flexible, and interpretable fits are desired. We propose the (FLAM), in which each additive function is estimated to be piecewise constant with a small number of adaptively-chosen knots. FLAM is the solution to a convex optimization problem, for which a simple algorithm with guaranteed convergence to a global optimum is provided. FLAM is shown to be consistent in high dimensions, and an unbiased estimator of its degrees of freedom is proposed. We evaluate the performance of FLAM in a simulation study and on two data sets. Supplemental materials are available online, and the R package flam is available on CRAN.
摘要
我们考虑在需要进行加法、灵活且可解释拟合的情况下，使用在独立观测值上测量的协变量来预测结果变量的问题。我们提出了灵活加法模型（FLAM），其中每个加法函数被估计为具有少量自适应选择节点的分段常数。FLAM是一个凸优化问题的解，为此提供了一种保证收敛到全局最优的简单算法。结果表明，FLAM在高维情况下是一致的，并且提出了其自由度的无偏估计量。我们在模拟研究和两个数据集上评估了FLAM的性能。补充材料可在线获取，R包flam可在CRAN上获取。
相似文献
1
Fused Lasso Additive Model.融合套索加法模型
J Comput Graph Stat. 2016;25(4):1005-1025. doi: 10.1080/10618600.2015.1073155. Epub 2016 Nov 10.
2
Convex Modeling of Interactions with Strong Heredity.具有强遗传性的相互作用的凸模型
J Comput Graph Stat. 2016;25(4):981-1004. doi: 10.1080/10618600.2015.1067217. Epub 2015 Aug 12.
3
Convex Regression with Interpretable Sharp Partitions.具有可解释尖锐划分的凸回归
J Mach Learn Res. 2016 Jun;17.
4
Data-adaptive additive modeling.数据自适应加法建模。
Stat Med. 2019 Feb 20;38(4):583-600. doi: 10.1002/sim.7859. Epub 2018 Jul 16.
5
CALIBRATING NON-CONVEX PENALIZED REGRESSION IN ULTRA-HIGH DIMENSION.超高维情形下非凸惩罚回归的校准
Ann Stat. 2013 Oct 1;41(5):2505-2536. doi: 10.1214/13-AOS1159.
6
Continuation of Nesterov's Smoothing for Regression With Structured Sparsity in High-Dimensional Neuroimaging.高维神经影像中具有结构化稀疏性的回归问题的 Nesterov 平滑的延续
IEEE Trans Med Imaging. 2018 Nov;37(11):2403-2413. doi: 10.1109/TMI.2018.2829802. Epub 2018 Apr 24.
7
Group Sparse Additive Models.组稀疏加法模型
Proc Int Conf Mach Learn. 2012;2012:871-878.
8
Controlling false discoveries in high-dimensional situations: boosting with stability selection.在高维情形下控制错误发现：基于稳定性选择的增强方法
BMC Bioinformatics. 2015 May 6;16:144. doi: 10.1186/s12859-015-0575-3.
9
Nonparametric regression with adaptive truncation via a convex hierarchical penalty.通过凸分层惩罚进行自适应截断的非参数回归
Biometrika. 2019 Mar;106(1):87-107. doi: 10.1093/biomet/asy056. Epub 2018 Dec 13.
10
Reader reaction to "Outcome-adaptive lasso: Variable selection for causal inference" by Shortreed and Ertefaie (2017).读者对 Shortreed 和 Ertefaie（2017）的“适应性lasso 方法：因果推断的变量选择”一文的反应。
Biometrics. 2023 Mar;79(1):514-520. doi: 10.1111/biom.13683. Epub 2022 May 31.
引用本文的文献
1
Covariance Assisted Multivariate Penalized Additive Regression (CoMPAdRe).协方差辅助多元惩罚加法回归（CoMPAdRe）。
J Comput Graph Stat. 2025;34(2):591-600. doi: 10.1080/10618600.2024.2407453. Epub 2024 Nov 22.
2
Multimodal Classification of Alzheimer's Disease Using Longitudinal Data Analysis and Hypergraph Regularized Multi-Task Feature Selection.基于纵向数据分析和超图正则化多任务特征选择的阿尔茨海默病多模态分类
Bioengineering (Basel). 2025 Apr 5;12(4):388. doi: 10.3390/bioengineering12040388.
3
A RELUCTANT ADDITIVE MODEL FRAMEWORK FOR INTERPRETABLE NONLINEAR INDIVIDUALIZED TREATMENT RULES.一种用于可解释非线性个体化治疗规则的非自愿加法模型框架。
Ann Appl Stat. 2023 Dec;17(4):3384-3402. doi: 10.1214/23-aoas1767. Epub 2023 Oct 30.
4
Adversarial Monte Carlo Meta-Learning of Optimal Prediction Procedures.最优预测程序的对抗蒙特卡罗元学习
J Mach Learn Res. 2021;22.
5
Generalized Sparse Additive Models.广义稀疏加性模型
J Mach Learn Res. 2022 Jan-Dec;23.
6
Reluctant Generalised Additive Modelling.非自愿广义相加模型
Int Stat Rev. 2020 Dec;88(Suppl 1):S205-S224. doi: 10.1111/insr.12429. Epub 2020 Nov 22.
7
Combining Metabolomics and Experimental Evolution Reveals Key Mechanisms Underlying Longevity Differences in Laboratory Evolved Populations.代谢组学与实验进化相结合揭示了实验室进化群体中长寿差异的关键机制。
Int J Mol Sci. 2022 Jan 19;23(3):1067. doi: 10.3390/ijms23031067.
8
Flexible and Interpretable Models for Survival Data.生存数据的灵活且可解释模型
J Comput Graph Stat. 2019;28(4):954-966. doi: 10.1080/10618600.2019.1592758. Epub 2019 May 20.
9
Adaptive nonparametric regression with the -nearest neighbour fused lasso.带 - 最近邻融合套索的自适应非参数回归。
Biometrika. 2020 Jun;107(2):293-310. doi: 10.1093/biomet/asz071. Epub 2020 Jan 29.
10
Nonparametric regression with adaptive truncation via a convex hierarchical penalty.通过凸分层惩罚进行自适应截断的非参数回归
Biometrika. 2019 Mar;106(1):87-107. doi: 10.1093/biomet/asy056. Epub 2018 Dec 13.
本文引用的文献
1
STANDARDIZATION AND THE GROUP LASSO PENALTY.标准化与组套索惩罚
Stat Sin. 2012 Jul;22(3):983-1001. doi: 10.5705/ss.2011.075.
2
A mathematical model for interpretable clinical decision support with applications in gynecology.一种具有可解释性的临床决策支持的数学模型及其在妇科的应用。
PLoS One. 2012;7(3):e34312. doi: 10.1371/journal.pone.0034312. Epub 2012 Mar 29.
3
Linear or Nonlinear? Automatic Structure Discovery for Partially Linear Models.线性还是非线性？部分线性模型的自动结构发现
J Am Stat Assoc. 2011 Sep 1;106(495):1099-1112. doi: 10.1198/jasa.2011.tm10281.
4
Autism and increased paternal age related changes in global levels of gene expression regulation.自闭症与父代年龄增加相关的基因表达调控的全球水平变化。
PLoS One. 2011 Feb 17;6(2):e16715. doi: 10.1371/journal.pone.0016715.
5
VARIABLE SELECTION IN NONPARAMETRIC ADDITIVE MODELS.非参数加法模型中的变量选择
Ann Stat. 2010 Aug 1;38(4):2282-2313. doi: 10.1214/09-AOS781.
6
Identification of a novel biomarker, SEMA5A, for non-small cell lung carcinoma in nonsmoking women.鉴定一种新型生物标志物 SEMA5A 用于非吸烟女性的非小细胞肺癌。
Cancer Epidemiol Biomarkers Prev. 2010 Oct;19(10):2590-7. doi: 10.1158/1055-9965.EPI-10-0332. Epub 2010 Aug 27.
7
Airway epithelial gene expression in the diagnostic evaluation of smokers with suspect lung cancer.气道上皮基因表达在疑似肺癌吸烟者诊断评估中的作用
Nat Med. 2007 Mar;13(3):361-6. doi: 10.1038/nm1556. Epub 2007 Mar 4.
8
NCBI GEO: mining tens of millions of expression profiles--database and tools update.NCBI基因表达综合数据库：挖掘数千万个表达谱——数据库与工具更新
Nucleic Acids Res. 2007 Jan;35(Database issue):D760-5. doi: 10.1093/nar/gkl887. Epub 2006 Nov 11.
9
Generalized additive models for medical research.医学研究中的广义相加模型。
Stat Methods Med Res. 1995 Sep;4(3):187-96. doi: 10.1177/096228029500400302.
Zotero 插件