Local Minimax Complexity of Stochastic Convex Optimization

来源 :上海交通大学 | 被引量 : 0次 | 上传用户：fa2009

【摘要】

：

We extend the traditional worst-case,minimax analysis of stochastic convex optimization by introducing a localized form of minimax complexity for individual functions.

【作者】

：

Sabyasachi Chatterjee John Duchi John Lafferty Yuancheng Zhu

【机构】

：

University of Chicago,USA Stanford University,USA

【出处】

：

上海交通大学

【发表日期】

：

2016年4期

【关键词】

：

Minimax optimality hardest local alternative stochastic optimization

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 extend the traditional worst-case,minimax analysis of stochastic convex optimization by introducing a localized form of minimax complexity for individual functions.

其他文献

BAYESIAN METHOD AND ITS APPLICATION TO CHINESE YIELD CURVE ESTIMATION

The term structure of interest rates,also called the yield curve,describes how the interest rate of bonds evolves over time,which is a central concept in economic and financial theory.

会议

Bayesian methodYield curveBond valuationBayesian smoothing spline

Blocking DSDs and DSD-based Mixed Level Screening Designs

Screening designs are designs conducted at the primary stage of the experiment to screen out significant factors from a large number of factors for further studies.

会议

D-efficiencyInterchange algorithmNoise factorsOrthogonal blockingRow-Column

Genomic modelling for cancer survival prediction

For clinical genomic studies with high-dimensional datasets,tree-based ensemble methods offer a powerful solution for variable selection and prediction taking into account the complex interrelationshi

会议

OncogenomicHigh-dimensional dataSurvival treeNonsusceptible individualsBaggi

Goodness-of-fit Test for Distribution Function in the Presence of Recurrent Event Data

会议

Leveraging Mixed and Incomplete Outcomes via A Generalized Reduced Rank Regression

Large-scale predictive modeling tasks routinely emerge in various fields.In many realworld problems,the collected outcomes are of mixed types,including continuous measurements,binary indicators and co

会议

incomplete dataintegrative analysislow rank matrixmixed outcomesmultivariate

Persistence of centrality in random growing trees

We investigate properties of node centrality in random growing tree models.We focus on a measure of centrality that computes the maximum subtree size of the tree rooted at each node,with the most cent

会议

Random treesbranching processescentrality

Modeling dependent functional data

Functional data vectors consist of samples of multivariate data in which each component is a random function.

会议

Covariance modelingFunctional vectorsLongitudinal dataMultivariate stochastic

Optimal Fingerprinting in Detection and Attribution of Changes in Climate Extremes with Combined Sco

Detection and attribution in climate science aims to determine whether certain aspect of climate has changed over time and subsequently attribute the change to causes such as human activities.

会议

Estimating functionExtreme value analysisIndependence likelihoodSpatial depen

Stein's Method for Steady-State Diffusion Approximations

We consider the problem of approximating the number of customers in the steady state of a queuing system by the stationary distribution of a diffusion process.

会议

Stein's methodKolmogorov boundmoderate deviationqueuing model

Peter Hall's Contribution to Empirical Likelihood

会议

Local Minimax Complexity of Stochastic Convex Optimization

其他学术论文