检索结果-内蒙古大学图书馆

On the computation and efficiency of a HBP-GM estimator some simulation results

COMPUTATIONAL STATISTICS & DATA ANALYSIS 1997年第1期25卷 1-15页

作者： Hinloopen, J Wagenvoort, R EUROPEAN UNIV INST DEPT ECON I-50016 FIESOLE FI ITALY

We propose and test a specific correction factor which improves both the resampling and projection algorithm for approximating the minimum volume ellipsoid (MVE) estimator. Simulations show that a high-breakdown-point GM estimator, based among other things on these improved MVE-estimates of location and scatter (i) is little less efficient than OLS if data are free from outlying observations, but in most cases is much more efficient if outliers corrupt the data, (ii) is always more efficient than Rousseeuw's least median of squares (LMS) estimator, and (iii) is always superior to both LMS and OLS if both precision and efficiency are considered. (C) 1997 Elsevier Science B.V.

关键词： LMS MVE resampling algorithm projection algorithm efficiency

来源：评论

学校读者我要写书评

暂无评论

Parallel algorithms and Concentration Bounds for the Lovasz Local Lemma via Witness DAGs

引用

ACM TRANSACTIONS ON algorithmS 2017年第4期13卷 1–25页

作者： Haeupler, Bernhard Harris, David G. Carnegie Mellon Univ Sch Comp Sci Pittsburgh PA 15213 USA Univ Maryland Dept Comp Sci College Pk MD 20742 USA

The Lovasz Local Lemma (LLL) is a cornerstone principle in the probabilistic method of combinatorics, and a seminal algorithm of Moser and Tardos (2010) provides an efficient randomized algorithm to implement it. This can be parallelized to give an algorithm that uses polynomially many processors and runs in O(log(3) n) time on an EREW PRAM, stemming from O(log n) adaptive computations of a maximal independent set (MIS). Chung et al. (2014) developed faster local and parallel algorithms, potentially running in time O(log(2) n), but these algorithms require more stringent conditions than the LLL. We give a new parallel algorithm that works under essentially the same conditions as the original algorithm of Moser and Tardos but uses only a single MIS computation, thus running in O(log(2) n) time on an EREW PRAM. This can be derandomized to give an NC algorithm running in time O(log(2) n) as well, speeding up a previous NC LLL algorithm of Chandrasekaran et al. (2013). We also provide improved and tighter bounds on the runtimes of the sequential and parallel resampling-based algorithms originally developed by Moser and Tardos. These apply to any problem instance in which the tighter Shearer LLL criterion is satisfied.

关键词： Lovasz Local Lemma Shearer's criterion resampling algorithm

来源：评论

学校读者我要写书评

暂无评论

Predicting Multiple Outcomes Associated with Frailty based on Imbalanced Multi-label Classification

引用

JOURNAL OF HEALTHCARE INFORMATICS RESEARCH 2024年第4期8卷 594-618页

作者： Tarekegn, Adane Nega Michalak, Krzysztof Costa, Giuseppe Ricceri, Fulvio Giacobini, Mario Univ Bergen Dept Informat Sci & Media Studies Bergen Norway Wroclaw Univ Econ & Business Dept Informat Technol Wroclaw Poland Univ Turin Dept Clin & Biol Sci Turin Italy Bahir Dar Univ Bahir Dar Inst Technol Fac Comp Bahir Dar Ethiopia Univ Turin Dept Vet Sci Data Anal & Modeling Unit Turin Italy

Frailty syndrome is prevalent among the elderly, often linked to chronic diseases and resulting in various adverse health outcomes. Existing research has predominantly focused on predicting individual frailty-related outcomes. However, this paper takes a novel approach by framing frailty as a multi-label learning problem, aiming to predict multiple adverse outcomes simultaneously. In the context of multi-label classification, dealing with imbalanced label distribution poses inherent challenges to multi-label prediction. To address this issue, our study proposes a hybrid resampling approach tailored for handling imbalance problems in the multi-label scenario. The proposed resampling technique and prediction tasks were applied to a high-dimensional real-life medical dataset comprising individuals aged 65 years and above. Several multi-label algorithms were employed in the experiment, and their performance was evaluated using multi-label metrics. The results obtained through our proposed approach revealed that the best-performing prediction model achieved an average precision score of 83%. These findings underscore the effectiveness of our method in predicting multiple frailty outcomes from a complex and imbalanced multi-label dataset.

关键词： Frailty prediction Hybrid resampling Imbalanced data Multi-label classification resampling algorithm

来源：评论

学校读者我要写书评

暂无评论

The Moser-Tardos Framework with Partial resampling

引用

JOURNAL OF THE ACM 2019年第5期66卷 36-36页

作者： Harris, David G. Srinivasan, Aravind Univ Maryland Dept Comp Sci AV Williams Bldg8223 Paint Branch Dr College Pk MD 20742 USA Univ Maryland Inst Adv Comp Studies College Pk MD 20742 USA

The resampling algorithm of Moser and Tardos is a powerful approach to develop constructive versions of the Lovasz Local Lemma. We generalize this to partial resampling: When a bad event holds, we resample an appropriately random subset of the variables that define this event rather than the entire set, as in Moser and Tardos. This is particularly useful when the bad events are determined by sums of random variables. This leads to several improved algorithmic applications in scheduling, graph transversals, packet routing, and so on. For instance, we settle a conjecture of Szabo and Tardos (2006) on graph transversals asymptotically and obtain improved approximation ratios for a packet routing problem of Leighton, Maggs, and Rao (1994).

关键词： Lovasz local lemma resampling algorithm discrepancy packet routing independent transversal

来源：评论

学校读者我要写书评

暂无评论

resampling Method to Compensate the Predicted Error Induced by Spectral Shift for Partial Least Squares (PLS) Models in Laser Raman Spectroscopy

引用

LASERS IN ENGINEERING 2020年第4-6期45卷 283-292页

作者： Bian, H. Wang, J. Yu, Y. S. Wang, X. Y. Gao, J. Huaiyin Inst Technol Fac Elect Informat Engn Huaian 223003 Jiangsu Peoples R China Chinese Acad Sci Suzhou Inst Biomed Engn & Technol Jiangsu Key Lab Med Opt Suzhou 215163 Jiangsu Peoples R China Suzhou ZeDaXingBang Pharmaceut Co Ltd Suzhou 215163 Jiangsu Peoples R China Szhou Guoke Med Sci & Technol Dev Co Ltd Suzhou 215163 Jiangsu Peoples R China

We propose herein an algorithm to resample the Raman spectra which aims to correct the predicted error induced by spectral shift. The same set of training data for both human and nonhuman blood Raman spectra was applied to build the model using partial least squares (PLS) method. Several kinds of Raman spectra of the blood samples originating from human, goose, monkey, chicken, sheep and duck without spectral shift were used to validate the model. The Raman spectra of the same blood samples with a shift of 9.18 cm(-1) were used to illustrate the influence of the spectral shift to the model. To demonstrate the effectiveness of the algorithm proposed, Raman spectra with spectral shift were resampled. The results show that the resampling algorithm can improve both the stability and accuracy of the model and reduce the impact of the spectral shift, which is promising for quantitative or qualitative analysis in Raman spectroscopy for complex samples.

关键词： Diode laser Raman spectroscopy partial least squares (PLS) spectral shift resampling algorithm blood discrimination

来源：评论

学校读者我要写书评

暂无评论

Computings Estimators For Regression And Multivariate Location/Dispersion

引用

Journal of Computational and Graphical Statistics 1992年第3期1卷 253-270页

作者： Ruppert, David School of Operations Research and Industrial Engineering Cornell University Ithaca NY 14850 United States

An improved resampling algorithm for S estimators reduces the number of times the objective function is evaluated and increases the speed of convergence. With this algorithm, S estimates can be computed in less time than least median squares (LMS) for regression and minimum volume ellipsoid (MVE) for location/scatter estimates with the same accuracy. Here accuracy refers to the randomness due to the algorithm. S estimators are also more statistically efficient than the LMS and MVE estimators, that is, they have less variability due to the randomness of the data. © 1992 Taylor & Francis Group, LLC.

关键词： Breakdown point Least median of squares estimator Minimum volume ellipsoid estimator Outlier detection Random direction search resampling algorithm Robust regression

来源：评论

学校读者我要写书评

暂无评论

MIMO Fingerprinting-Based Particle Filter for Mobile Positioning Systems

MIMO Fingerprinting-Based Particle Filter for Mobile Positio...

引用

International Symposium on VLSI Design, Automation, and Test (VLSI-DAT)

作者： Chuang, Mu-Hsuan Lo, Yi-Hao Wu, Bo-Yi Huang, Yuan-Hao Natl Tsing Hua Univ Inst Commun Engn Hsinchu Taiwan

ISBN: (纸本)9781467344364;9781467344357

This paper presents a particle filter processor for mobile positioning systems. The proposed particle filter employs a selective sampling scheme to reduce the required iteration of particle filtering and improve positioning accuracy. Moreover, a threshold IMH resampling algorithm is also proposed to enhance the hardware utilization rate and processing latency. This study also designs and implements the particle filering processor with a flexible particle number and sampling threshold according different channel conditions. The proposed threshold IMH resampling not only improves the hardware utilization to more than 95% but also reduces the processing latency of the conventional IMH resampling algorithm. The chip implementation results show that the particle filter, when deployed in base-station, can track the location of 100 mobile users at 180 km/hr speed with less than 10-meter root-mean-square (RMS) error.

关键词： Particle filters resampling algorithm BASE STATION PROCESSOR multiple input multiple output systems

来源：评论

学校读者我要写书评

暂无评论

An Effective FastSLAM algorithm Based on CUDA

引用

INTERNATIONAL JOURNAL OF GRID AND DISTRIBUTED COMPUTING 2016年第12期9卷 143-158页

作者： Zhang, Heng Liu, Yanli Zhu, Mengyu Xiong, Naixue Kim, Tai-hoon East China Jiaotong Univ Sch Informat Engn Nanchang Jiangxi Peoples R China Hubei Univ Educ Sch Comp Wuhan Peoples R China Southwestern Oklahoma State Univ Dept Business & Comp Sci Weatherford OK USA Hannam Univ Dept Multimedia Engn Daejeon South Korea

Compute Unified Device Architecture (CUDA) is a mature parallel computing architecture, which can significantly accelerate performance of the computation intensive algorithm. In this paper, FastSLAM algorithm based on the probability model is further studied and the resampling algorithm for the path estimation is improved. In the resampling phase, resampling rules are redesigned and the previous data limitations are broken for the purpose of parallelization. We propose the FastSLAM algorithm based on CUDA, which accelerates robot localization and mapping. The experiment results show that FastSLAM_CUDA can achieve a significant speedup over the FastSLAM with many particles.

关键词： compute unified device architecture (CUDA) simultaneous localization and mapping (SLAM) resampling algorithm parallel computing

来源：评论

学校读者我要写书评

暂无评论

Particle Filter for Positioning Accuracy Improvement in GNSS Receiver

Particle Filter for Positioning Accuracy Improvement in GNSS...

引用

2013 3rd International Conference on Electric and Electronics(EEIC 2013)

作者： Xiaowen Sun Shufang Zhang Qing Hu College of Information Science and Technology Dalian Maritime University

Traditional GNSS receiver usually uses Least Squares(LS) method or Kalman filtering to calculate the position. This paper proposes a new method that uses particle filtering for positioning calculation. The proposed particle filtering bases on LS method, but resolves the low positioning accuracy problem of LS. It also overcomes the Kalman filtering's drawback that need to know noise properties and receiver's dynamic model previously. The paper gives detailed implementation of the particle filter that is applicable to positioning calculation in GNSS receiver. Based on a suitably built model, the filter uses a modified resampling algorithm, and selects reasonable particle number and proposal distribution according to simulation test. These effectively reduce the calculation and improve the performance of the filter. Simulation results with GSS6700 simulator, based on the digital Intermediate Frequency(IF) signals, show that compared with traditional LS method and Kalman filtering, the proposed particle filtering achieves much higher positioning accuracy.

关键词： GNSS receiver Positioning accuracy Kalman filtering Particle filtering resampling algorithm

来源：评论

学校读者我要写书评

暂无评论

A method for automatic generation of caricatured profile in PICASSO system

引用

Systems and Computers in Japan 1998年第1期29卷 12-19页

作者： Hayashi, Jun-Ichiro Murakami, Kazuhito Koshimizu, Hiroyasu Sch. of Comp. and Cognitive Sciences Chukyo University Toyota 470-03 Japan Chukyo University Japan IECE Japan Tokai Al Study Group Japan Nagoya University Japan Nagotas Indust. Res. Laboratories Institute of Information Processing Japan Institute of Artificial Intelligence Japan Institute of Non-Distraction Tests Japan Institute of Facial Science Japan Yamanashi University Japan Nagoya-City Indust. Res. Labs. Japan Inst. of Measurements and Automation Japan Institute of Image Information Media Japan

In the generation of a caricatured profile of a human face, it is most important to recognize the contour of each facial part. Notably, in the extraction of personal features from the portrait and their caricatured exaggeration by using an "average profile" and extrapolation, the correspondence between the contours of two types of portrait is crucial. The authors have previously proposed a method of automatic generation of caricatured profiles, by using a simple contour extraction, a hierarchical resampling, and the deformation of a facial contour and other facial parts which is a combination of extrapolation and the affine transform of partial regions. The previous method can be processed automatically, except for two manual processes of normalization. This paper examines various points of the previous method so that a totally automatic processing is possible. © 1998 Scripta Technica.

关键词： Affine transform Caricatured profile Feature extraction resampling algorithm

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：