发布求助
文献互助 智能选刊 最新文献

2018 IEEE Statistical Signal Processing Workshop (SSP)最新文献

筛选
英文 中文
Combining SVMS for Classification on Class Imbalanced Data 结合支持向量机的类不平衡数据分类
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450746
S. Sukhanov, A. Merentitis, C. Debes, Jürgen T. Hahn, A. Zoubir
{"title":"Combining SVMS for Classification on Class Imbalanced Data","authors":"S. Sukhanov, A. Merentitis, C. Debes, Jürgen T. Hahn, A. Zoubir","doi":"10.1109/SSP.2018.8450746","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450746","url":null,"abstract":"The class imbalance problem in classification scenarios is considered to be one of the main issues that limits the performance of many learning techniques. When reporting high classification accuracy a classifier may still exhibit poor performance for the minority class that is often the class of interest. In this paper, we propose to address the class imbalance problem by applying an SVM-based ensemble framework that provides the ability to control the trade-off between discovery rate of the underrepresented classes and the overall accuracy simultaneously. We evaluate the performance of the proposed technique on both synthetic and real-world datasets demonstrating the advantage of the method compared to state-of-the-art approaches.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121606330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Minimax Lower Bounds for Nonnegative Matrix Factorization 非负矩阵分解的极大极小下界
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450822
Mine Alsan, Zhaoqiang Liu, V. Tan
{"title":"Minimax Lower Bounds for Nonnegative Matrix Factorization","authors":"Mine Alsan, Zhaoqiang Liu, V. Tan","doi":"10.1109/SSP.2018.8450822","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450822","url":null,"abstract":"The non-negative matrix factorization (NMF) problem consists in modeling data samples as non-negative linear combinations of non-negative dictionary vectors. While many algorithms for NMF have been proposed, fundamental performance limits of these algorithms are currently not available. This paper plugs this gap by providing lower bounds on the minimax risk (the minimum achievable worst case mean squared error) of estimating the non-negative dictionary matrix under a set of locality and statistical assumptions.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116156030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Multipath Mitigation in Global Navigation Satellite Systems Using a Bayesian Hierarchical Model With Bernoulli Laplacian Priors 基于贝叶斯层次模型和伯努利拉普拉斯先验的全球卫星导航系统多径缓解
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450818
Julien Lesouple, J. Tourneret, M. Sahmoudi, Franck Barbiero, Frederic Faurie
{"title":"Multipath Mitigation in Global Navigation Satellite Systems Using a Bayesian Hierarchical Model With Bernoulli Laplacian Priors","authors":"Julien Lesouple, J. Tourneret, M. Sahmoudi, Franck Barbiero, Frederic Faurie","doi":"10.1109/SSP.2018.8450818","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450818","url":null,"abstract":"A new sparse estimation method was recently introduced in a previous work to correct biases due to multipath (MP) in GNSS measurements. The proposed strategy was based on the resolution of a LASSO problem constructed from the navigation equations using the reweighted $-ell _{1}$ method. This strategy requires to adjust the regularization parameters balancing the data fidelity term and the involved regularizations. This paper introduces a new Bayesian estimation method allowing the MP biases and the unknown model parameters and hyperparameters to be estimated directly from the GNSS measurements. The proposed method is based on Bernoulli-Laplacian priors, promoting sparsity of MP biases.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116560380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
The Geometry of Coherence and Its Application to Cyclostationary Time Series 相干几何及其在周期平稳时间序列中的应用
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450812
S. Howard, S. Sirianunpiboon, D. Cochran
{"title":"The Geometry of Coherence and Its Application to Cyclostationary Time Series","authors":"S. Howard, S. Sirianunpiboon, D. Cochran","doi":"10.1109/SSP.2018.8450812","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450812","url":null,"abstract":"The consequences of cyclostationary structure in a random process have traditionally been described in terms of the correlation or coherence of pairs of particular time and frequency shifted versions of the process. However, cyclostationarity, and more generally almost cyclostationarity, are manifest in the mutual coherence of subspaces spanned by sets of time and frequency shifted versions of the process. The generalized coherence framework allows any finite collection of pertinent samples of the cyclic autocorrelation function estimates formed from the measured signal data to be combined into a detection statistic. This paper develops the subspace coherence theory of almost cyclostationary processes as a guide to constructing such detectors in both the time and spectral domains.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115789075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
Random Hyperplanes, Generalized Singular Values & “What’s My β?” 随机超平面、广义奇异值与“我的β是什么?”
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450833
A. Edelman, Bernie Wang
{"title":"Random Hyperplanes, Generalized Singular Values & “What’s My β?”","authors":"A. Edelman, Bernie Wang","doi":"10.1109/SSP.2018.8450833","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450833","url":null,"abstract":"We streamline the treatment of the Jacobi ensemble from random matrix theory by providing a succinct geometric characterization which may be used directly to compute the Jacobi ensemble distribution without unnecessary matrix baggage traditionally seen in the MANOVA formulation. Algebraically the Jacobi ensemble naturally corresponds to the Generalized Singular Value Decomposition from the field of Numerical Linear Algebra. We further provide a clear geometric interpretation for the Selberg constant in front of the distribution which may sensibly be defined even beyond the reals, complexes, and quaternions. On the application side, we propose a new learning problem where one estimates a $beta $ that best fits the sample eigenvalues from the Jacobi ensemble.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125858838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Differential Privacy for Positive and Unlabeled Learning With Known Class Priors 具有已知类先验的正和无标记学习的差分隐私
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450839
Anh T. Pham, R. Raich
{"title":"Differential Privacy for Positive and Unlabeled Learning With Known Class Priors","authors":"Anh T. Pham, R. Raich","doi":"10.1109/SSP.2018.8450839","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450839","url":null,"abstract":"Despite the increasing attention to big data, there are several domains where labeled data is scarce or too costly to obtain. For example, for data from information retrieval, gene analysis, and social network analysis, only training samples from the positive class are annotated while the remaining unlabeled training samples consist of both unlabeled positive and unlabeled negative samples. The specific positive and unlabeled (PU) data from those domains necessitates a mechanism to learn a two-class classifier from only one-class labeled data. Moreover, because data from those domains is highly sensitive and private, preserving training samples privacy is essential. This paper addresses the challenge of private PU learning by designing a differentially private algorithm for positive and unlabeled data. We first propose a learning framework for the PU setting when the class prior probability is known, with a theoretical guarantee of convergence to the optimal classifier. We then propose a privacy-preserving mechanism for the designed framework where the privacy and utility are both theoretically and empirically proved.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122281839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
A Double Proposal Normalized Importance Sampling Estimator 一种双建议归一化重要抽样估计
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450849
Roland Lamberti, Y. Petetin, F. Septier, F. Desbouvries
{"title":"A Double Proposal Normalized Importance Sampling Estimator","authors":"Roland Lamberti, Y. Petetin, F. Septier, F. Desbouvries","doi":"10.1109/SSP.2018.8450849","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450849","url":null,"abstract":"Monte Carlo methods are widely used in signal processing for computing integrals of interest. Among Monte Carlo methods, Importance Sampling is a variance reduction technique which consists in sampling from an instrumental distribution and reweighting the samples in order to correct the discrepancy between the target and proposal distributions. When either the target or the proposal distribution is known only up to a constant, the moment of interest can be rewritten as a ratio of two expectations, which can be approximated via self-normalized importance sampling. In this paper we show that it is possible to improve the self-normalized importance sampling estimate by approximating the two expectations in this ratio via two importance distributions. In order to tune them we optimize the variance of the final estimate under a reasonable constraint. Our results are validated via simulations.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128143911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Distributed Wiener-Based Reconstruction of Graph Signals 基于分布式维纳的图信号重构
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450828
E. Isufi, P. Lorenzo, P. Banelli, G. Leus
{"title":"Distributed Wiener-Based Reconstruction of Graph Signals","authors":"E. Isufi, P. Lorenzo, P. Banelli, G. Leus","doi":"10.1109/SSP.2018.8450828","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450828","url":null,"abstract":"This paper proposes strategies for distributed Wiener-based reconstruction of graph signals from subsampled measurements. Given a stationary signal on a graph, we fit a distributed autoregressive moving average graph filter to a Wiener graph frequency response and propose two reconstruction strategies: i) reconstruction from a single temporal snapshot; ii) recursive signal reconstruction from a stream of noisy measurements. For both strategies, a mean square error analysis is performed to highlight the role played by the filter response and the sampled nodes, and to propose a graph sampling strategy. Our findings are validated with numerical results, which illustrate the potential of the proposed algorithms for distributed reconstruction of graph signals.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128210842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 16
Joint Sensing Matrix Design And Recovery Based On Normalized Iterative Hard Thesholding for Sparse Systems 基于归一化迭代硬分割的稀疏系统联合感知矩阵设计与恢复
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450696
Qianru Jiang, R. D. Lamare, Y. Zakharov, Sheng Li, Xiongxiong He
{"title":"Joint Sensing Matrix Design And Recovery Based On Normalized Iterative Hard Thesholding for Sparse Systems","authors":"Qianru Jiang, R. D. Lamare, Y. Zakharov, Sheng Li, Xiongxiong He","doi":"10.1109/SSP.2018.8450696","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450696","url":null,"abstract":"In this work, we present a joint sensing matrix design and recovery algorithm based on the normalized iterative hard thresholding (NIHT) algorithm for cost-effectively solving the problem of sparse recovery. In particular, we consider both the Gram of the sensing matrix and a gradient-based algorithm based on the real mutual coherence (RMC) to compute the sensing matrix, so that the Gram of the matrix can closely approach the relaxed equiangular tight frame (ETF. By optimizing the sensing matrix together with its column normalization, a better recovery performance can be achieved. Simulations assess the performance of the proposed approach versus other iterative hard thresholding-based algorithms and show that the proposed approach achieves the best recovery performance.","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131481069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 3
Random Matrix-Improved Kernels For Large Dimensional Spectral Clustering 大维谱聚类的随机矩阵改进核算法
2018 IEEE Statistical Signal Processing Workshop (SSP) Pub Date : 2018-06-01 DOI: 10.1109/SSP.2018.8450705
Hafiz Tiomoko Ali, A. Kammoun, Romain Couillet
{"title":"Random Matrix-Improved Kernels For Large Dimensional Spectral Clustering","authors":"Hafiz Tiomoko Ali, A. Kammoun, Romain Couillet","doi":"10.1109/SSP.2018.8450705","DOIUrl":"https://doi.org/10.1109/SSP.2018.8450705","url":null,"abstract":"Leveraging on recent random matrix advances in the performance analysis of kernel methods for classification and clustering, this article proposes a new family of kernel functions theoretically largely outperforming standard kernels in the context of asymptotically large and numerous datasets. These kernels are designed to discriminate statistical means and covariances across data classes at a theoretically minimal rate (with respect to data size). Applied to spectral clustering, we demonstrate the validity of our theoretical findings both on synthetic and real-world datasets (here, the popular MNIST database as well as EEG recordings on epileptic patients).","PeriodicalId":330528,"journal":{"name":"2018 IEEE Statistical Signal Processing Workshop (SSP)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131824051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 5
首页 上一页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：604180095
Book学术官方微信