Density Estimation in Infinite Dimensional Exponential Families

来源 :上海交通大学 | 被引量 : 0次 | 上传用户：tdcdc

【摘要】

：

【作者】

：

Bharath Sriperumbudur Kenji Fukumizu Arthur Gretton Aapo Hyv(a)rinen Revant Kumar

【机构】

：

Department of Statistics,Pennsylvania State University,University Park,PA 16802,USA

【出处】

：

上海交通大学

【发表日期】

：

2016年4期

【关键词】

：

density estimation exponential family Fisher divergence reproducing kernel Hilbe

下载到本地 , 更方便阅读

下载此文赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

　　We consider an infinite dimensional generalization of natural exponential family of probability densities,which are parametrized by functions in a reproducing kernel Hilbert space(RKHS),and show it to be quite rich in the sense that a broad class of densities on Rd can be approximated arbitrarily well in Kullback-Leibler(KL)divergence by elements in the infinite dimensional family P.

其他文献

POST-CHANGE INFERENCE AFTER SEQUENTIAL CHANGE-POINT DETECTION IN PANEL DATA

Assume a common change may occur in a portion of N independent panels and each panel consists of a sequence of independent random variables following a standard exponential family distribution.

会议

change-point problemexponential familylocal scoremixture modelShiryayev-Robe

A Time Course Inferential Framework for Gene Set Enrichment Analysis Based on Functional Elastic-net

Gene set enrichment analyses(GSEA)are powerful inferential methods widely used in genomic research to identify significant gene sets,such as Gene Ontology terms and molecular pathways.

会议

gene set enrichment analysisfunctional data analysiselastic-netnonparametric

Backward Nested Descriptors Asymptotics with Inference on Stem Cell Differentiation

For sequences of random backward nested subspaces as occur,say,in dimension reduction for manifold or stratified space valued data,asymptotic results are derived.

会议

Fr(e)chet meansdimension reduction on manifoldsprincipal nested spheresasympt

Adaptive Signal Inclusion With Genomic Applications

In ultrahigh dimensional data analysis,variable screening has been routinely per-formed before tting complex models with the goal to e ciently reduce data dimension while still keeping majority of th

会议

Combination Rule in the Rare Disease Domain(Design and Testing Strategy)

We discuss the design and testing strategy for a clinical trial involving Combination therapy(A+B),Monotherapy(A),and Placebo(P).

会议

rare diseaseBIBDcombination therapytesting strategytruncated Hochberg

Assessment of p-value quality in Phase 3 Clinical Trial

A p-value is a most widely used statistical measure of evidence against a null hypothesis in statistical hypothesis testing and is generally required in Phase 3 clinical trials for drug registration a

会议

p-valuemeasure of evidencereproducibilityprediction intervalclinical trial

LATENT-SCALE DISTANCE-BASED MODELS FOR RANK AGGREGATION

Rank aggregation aims at combining rankings of a set of items assigned by a sample of rankers to generate a better consensus ranking.

会议

Distance-basedRanking dataRank aggregation

Statistical Model Building for Large,Complex Data:New Directions in SAS/STAT(R)Software

The increasing size and complexity of data in research and business applications require a more versatile set of tools for building explanatory and predictive statistical models.

会议

softwaremodel selectionregressionprediction

A Bayesian Spatio-Temporal Factor Analysis Model for Predicting Sea Levels Along the US Coast

Monitoring sea levels at a local level is vital for protecting coastal populations,ecological systems,and natural resources.

会议

spatially-varying coefficientslarge datauncertainty quantification

Fiducial Inference on a Biomarker Adaptive Design

A clinical trial,adaptive design that allows pre-specified modifications in the design or statistical procedures of the trial depending on accrued data makes a trial more flexible.

会议

Clinical TrialBiomarkerAdoptive DesignFiducial Inference

Density Estimation in Infinite Dimensional Exponential Families

其他学术论文