检索结果-内蒙古大学图书馆

33rd Conference on Learning Theory (COLT)

作者： Avanesov, Valeriy WIAS Berlin Berlin Germany

Gaussian Process regression and Kernel Ridge regression are popular nonparametric regression approaches. Unfortunately, they suffer from high computational complexity rendering them inapplicable to the modern massive datasets. To that end a number of approximations have been suggested, some of them allowing for a distributed implementation. One of them is the divide and conquer approach, splitting the data into a number of partitions, obtaining the local estimates and finally averaging them. In this paper we suggest a novel computationally efficient fully data-driven algorithm, quantifying uncertainty of this method, yielding frequentist L-2-confidence bands. We rigorously demonstrate validity of the algorithm. Another contribution of the paper is a minimax-optimal high-probability bound for the averaged estimator, complementing the known risk bounds.

关键词： Gaussian process regression kernel ridge regression nonparametric regression distributed regression confidence bands bootstrap

来源：评论

学校读者我要写书评

暂无评论

Combining distributed regression and propensity scores: a doubly privacy-protecting analytic method for multicenter research

引用

CLINICAL EPIDEMIOLOGY 2018年 10卷 1773-1786页

作者： Toh, Sengwee Wellman, Robert Coley, R. Yates Horgan, Casie Sturtevant, Jessica Moyneur, Erick Janning, Cheri Pardee, Roy Coleman, Karen J. Arterburn, David McTigue, Kathleen Anau, Jane Cook, Andrea J. Harvard Med Sch Dept Populat Med 401 Pk DrSuite 401 East Boston MA 02215 USA Harvard Pilgrim Hlth Care Inst 401 Pk DrSuite 401 East Boston MA 02215 USA Kaiser Permanente Washington Hlth Res Inst Seattle WA USA StatLog Econometr Inc Montreal PQ Canada Duke Clin & Translat Sci Inst Durham NC USA Kaiser Permanente Southern Calif Pasadena CA USA Univ Pittsburgh Dept Med Pittsburgh PA USA

Purpose: Sharing of detailed individual-level data continues to pose challenges in multicenter studies. This issue can be addressed in part by using analytic methods that require only summary-level information to perform the desired multivariable-adjusted analysis. We examined the feasibility and empirical validity of 1) conducting multivariable-adjusted distributed linear regression and 2) combining distributed linear regression with propensity scores, in a large distributed data network. Patients and methods: We compared percent total weight loss 1-year postsurgery between Roux-en-Y gastric bypass and sleeve gastrectomy procedure among 43,110 patients from 36 health systems in the National Patient-Centered Clinical Research Network. We adjusted for baseline demographic and clinical variables as individual covariates, deciles of propensity scores, or both, in three separate outcome regression models. We used distributed linear regression, a method that requires only summary-level information (specifically, sums of squares and cross products matrix) from sites, to fit the three ordinary least squares linear regression models. A comparison set of analyses that used pooled deidentified individual-level data from sites served as the reference. Results: distributed linear regression produced results identical to those from the corresponding pooled individual-level data analysis for all variables in all three models. The maximum numerical difference in the parameter estimate or standard error for all the variables was 3x10(-11) across three models. Conclusion: distributed linear regression analysis is a feasible and valid analytic method in multicenter studies for one-time continuous outcomes. Combining distributed regression with propensity scores via modeling offers more privacy protection and analytic flexibility.

关键词： distributed regression propensity score distributed data networks privacy-protecting methods

来源：评论

学校读者我要写书评

暂无评论

distributed Nonlinear regression Using In-Network Processing With Multiple Gaussian Kernels 18

Distributed Nonlinear Regression Using In-Network Processing...

引用

18th IEEE International Workshop on Signal Processing Advances for Wireless Communications (SPAWC)

作者： Shin, Ban-Sok Paul, Henning Yukawa, Masahiro Dekorsy, Armin Univ Bremen Dept Commun Engn Bremen Germany Keio Univ Dept Elect & Elect Engn Tokyo Japan

ISBN: (纸本)9781509030095

In this paper, we propose the use of multiple Gaussian kernels for distributed nonlinear regression or system identification tasks by a network of nodes. By employing multiple kernels in the estimation process we increase the degree of freedom and thus, the ability to reconstruct nonlinear functions. For this, we extend the so-called KDiCE algorithm, which allows a distributed regression of nonlinear functions but uses a single kernel only, to multiple kernels. We corroborate our proposed scheme by numerical evaluations for the reconstruction of nonlinear functions both static and time-varying. We achieve performance gains for both cases, in particular for the tracking of a time-varying nonlinear function.

关键词： In-network processing distributed regression kernel least-squares multiple kernels

来源：评论

学校读者我要写书评

暂无评论

In-cluster vector evaluated particle swarm optimization for distributed regression in WSNs

引用

JOURNAL OF NETWORK AND COMPUTER APPLICATIONS 2014年第Jun.期42卷 80-91页

作者： Shakibian, Hadi Charkari, Nasrollah Moghadam Tarbiat Modares Univ Fac Elect & Comp Engn Parallel Proc Lab 6 51 Tehran Iran

Conventional methods address data modeling in WSNs by converting a learning problem into an optimization task. These methods are mainly based on Gradient Descent and Nelder-Mead Simplex optimization techniques. They are energy efficient, but with drawbacks: the accuracy of the induced models is usually lower than that of the centralized technique;and the latency is proportional to the network size. In this paper, we propose a novel distributed method for parametric regression in a clustered WSN using particle swarm optimization. Two main contributions of this paper are as follows: First, we re-formulate the distributed regression in WSNs as a multiobjective optimization (MO) problem in which an objective is dedicated to each cluster. Second, we propose a distributed algorithm based on the Vector Evaluated Particle Swarm Optimization method to address the MO problem in two phases. The proposed algorithm obtains a set of candidate network regressors and computes the final model using a weighted averaging rule. We compare the prediction accuracy, latency, and energy consumption of our algorithm against its popular distributed counterparts and the centralized technique using a real-world and an artificial data set. The experimental results show that our algorithm outperforms the existing approaches in prediction accuracy while its energy consumption and latency are acceptable. We also evaluate the effect of the number of particles, swarm topologies, and clustering granularity on the performance of the proposed method. (C) 2014 Elsevier Ltd. All rights reserved.

关键词： distributed regression Particle swarm optimization Wireless sensor networks Multiobjective optimization distributed optimization

来源：评论

学校读者我要写书评

暂无评论

Auto-tuning procedures for distributed nonparametric regression algorithms

Auto-tuning procedures for distributed nonparametric regress...

引用

European Control Conference (ECC)

作者： Varagnolo, Damiano Pillonetto, Gianluigi Schenato, Luca Lulea Univ Technol Dept Comp Sci Elect & Space Engn Lulea Sweden Univ Padua Dept Informat Engn I-35100 Padua Italy

ISBN: (纸本)9783952426937

We propose a distributed regression algorithm with the capability of automatically calibrating its parameters during its on-line functioning. The estimation procedure corresponds to a Regularization Network, i.e., the structural form of the estimator is a linear combination of basis functions which coefficients are computed by solving a linear system. The automatic tuning strategy instead constructs and then exploits opportune bounds on the distance between the distributed estimation results and the unknown centralized optimal estimate that would be computed processing the whole dataset at once. By numerical simulations we show how the proposed procedure allows the sensor networks to effectively self-tune the parameters of the distributed regression scheme by simple consensus strategies.

关键词： distributed regression distributed calibration self-organizing sensor networks regularization networks non-parametric estimation

来源：评论

学校读者我要写书评

暂无评论

A Multi-Modal distributed Learning Algorithm in Reproducing Kernel Hilbert Spaces 6

A Multi-Modal Distributed Learning Algorithm in Reproducing ...

引用

6th Annual Learning for Dynamics and Control Conference

作者： Raghavan, Aneesh Johansson, Karl Henrik KTH Royal Inst Technol DCS Div Stockholm Sweden

We consider the problem of function estimation by a multi-agent system consisting of two agents and a fusion center. Each agent receives data comprising of samples of an independent variable (input) and the corresponding values of the dependent variable (output). The data remains local and is not shared with other members in the system. The objective of the system is to collaboratively estimate the function from the input to the output. To this end, we present an iterative distributed algorithm for this function estimation problem. Each agent solves a local estimation problem in a Reproducing Kernel Hilbert Space (RKHS) and uploads the function to the fusion center. At the fusion center, the functions are fused by first estimating the data points that would have generated the uploaded functions and then subsequently solving a least squares estimation problem using the estimated data from both functions. The fused function is downloaded by the agents and is subsequently used for estimation at the next iteration along with incoming data. This procedure is executed sequentially and stopped when the difference between consecutively estimated functions becomes small enough. With respect to the algorithm, we prove existence of basis functions for suitable representation of estimated functions and present closed form solutions to the estimation problems at the agents and the fusion center.

关键词： distributed regression RKHS Transfer Operators

来源：评论

学校读者我要写书评

暂无评论

Privacy-preserving estimation of an optimal individualized treatment rule: a case study in maximizing time to severe depression-related outcomes

引用

LIFETIME DATA ANALYSIS 2022年第3期28卷 512-542页

作者： Moodie, Erica E. M. Coulombe, Janie Danieli, Coraline Renoux, Christel Shortreed, Susan M. McGill Univ Dept Epidemiol Biostat & Occupat Hlth Montreal PQ Canada Jewish Gen Hosp Ctr Clin Epidemiol Lady Davis Inst Med Res Montreal PQ Canada McGill Univ Dept Neurol & Neurosurg Montreal PQ Canada Kaiser Permanente Washington Hlth Res Inst Biostat Unit Seattle WA USA Univ Washington Biostat Dept Seattle WA 98195 USA

Estimating individualized treatment rules-particularly in the context of right-censored outcomes-is challenging because the treatment effect heterogeneity of interest is often small, thus difficult to detect. While this motivates the use of very large datasets such as those from multiple health systems or centres, data privacy may be of concern with participating data centres reluctant to share individual-level data. In this case study on the treatment of depression, we demonstrate an application of distributed regression for privacy protection used in combination with dynamic weighted survival modelling (DWSurv) to estimate an optimal individualized treatment rule whilst obscuring individual-level data. In simulations, we demonstrate the flexibility of this approach to address local treatment practices that may affect confounding, and show that DWSurv retains its double robustness even when performed through a (weighted) distributed regression approach. The work is motivated by, and illustrated with, an analysis of treatment for unipolar depression using the United Kingdom's Clinical Practice Research Datalink.

关键词： Data aggregation distributed regression Dynamic weighted survival modelling Effect modification Precision medicine Selective serotonin reuptake inhibitors

来源：评论

学校读者我要写书评

暂无评论

Preserving data privacy when using multi-site data to estimate individualized treatment rules

引用

STATISTICS IN MEDICINE 2022年第9期41卷 1627-1643页

作者： Danieli, Coraline Moodie, Erica E. M. McGill Univ McGill Univ Hlth Ctr Res Inst Dept Epidemiol Biostat & Occupat Hlth Montreal PQ Canada McGill Univ Dept Epidemiol Biostat & Occupat Hlth Montreal PQ Canada

Precision medicine is a rapidly expanding area of health research wherein patient level information is used to inform treatment decisions. A statistical framework helps to formalize the individualization of treatment decisions that characterize personalized management plans. Numerous methods have been proposed to estimate individualized treatment rules that optimize expected patient outcomes, many of which have desirable properties such as robustness to model misspecification. However, while individual data are essential in this context, there may be concerns about data confidentiality, particularly in multi-center studies where data are shared externally. To address this issue, we compared two approaches to privacy preservation: (i) data pooling, which is a covariate microaggregation technique and (ii) distributed regression. These approaches were combined with the doubly robust yet user-friendly method of dynamic weighted ordinary least squares to estimate individualized treatment rules. In simulations, we extensively evaluated the performance of the methods in estimating the parameters of the decision rule under different assumptions. The results demonstrate that double robustness is not maintained in data pooling setting and that this can result in bias, whereas the distributed regression provides good performance. We illustrate the methods via an analysis of optimal Warfarin dosing using data from the International Warfarin Consortium.

关键词： data pooling distributed regression dynamic treatment regimes multi-Centre studies precision medicine

来源：评论

学校读者我要写书评

暂无评论

regression WITH AN ENSEMBLE OF NOISY BASE FUNCTIONS 32

REGRESSION WITH AN ENSEMBLE OF NOISY BASE FUNCTIONS

引用

IEEE 32nd International Workshop on Machine Learning for Signal Processing (MLSP)

作者： Ben-Hur, Yuval Cassuto, Yuval Cohen, Israel Technion Israel Inst Technol Viterbi Dept Elect & Comp Engn IL-3200003 Haifa Israel

ISBN: (数字)9781665485470

ISBN: (纸本)9781665485470

Ensemble methods achieve state-of-the-art performance in many real-world regression problems while enjoying structural compatibility for modern decentralized computing architectures. However, the implementation of ensemble regression on distributed systems may compromise its cutting-edge performance due to computing and communication reliability issues. This paper introduces robust ensemble combining techniques designed to integrate multiple noisy predictions into a single reliable prediction. Experiments conducted with synthetic and real-world datasets in various noise regimes illustrate our robust methods' superiority over non-robust counterparts.

关键词： Ensemble learning distributed regression inference noise

来源：评论

学校读者我要写书评

暂无评论

distributed cross-media multiple binary subspace learning

引用

INTERNATIONAL JOURNAL OF MULTIMEDIA INFORMATION RETRIEVAL 2015年第2期4卷 153-164页

作者： Zhao, Xueyi Zhang, Chenyi Zhang, Zhongfei Zhejiang Univ Dept Informat Sci & Elect Engn Hangzhou 310003 Zhejiang Peoples R China Simon Fraser Univ Sch Comp Sci Vancouver BC Canada SUNY Binghamton Dept Comp Sci Watson Sch Binghamton NY USA

Due to the ubiquitous existence of large-scale data in today's real-world applications, including learning on cross-media data, we propose a semi-supervised learning method, named Multiple Binary Subspace regression, for cross-media data concept detection. In order to mine the common features among the data with multiple modalities, we project the original cross-media data onto the same subspace-level representation simultaneously by mapping to the corresponding subspaces for dimensionality reduction. All the subspaces are set to be binary, which only involve the addition operations and omit the multiplication operations in the subsequent computation owing to the good property of the binary values. The dimensionality reduction to a binary subspace and the concept detection on this subspace are also optimized simultaneously leading to a semi-supervised model. For dealing with large-scale data, our learning method is easily implemented to run in a MapReduce-based Hadoop system. Empirical studies demonstrate its competitive performance on convergence, efficiency, and scalability in comparison with the state-ofthe- art literature.

关键词： distributed regression Binary subspace Cross media Parallel computation MapReduce

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：