The Stein Effect: an Alternative Film-induced Tourism Perspective

Graham Busby; Rong Huang; Rebecca Jarman

doi:10.1002/jtr.1875

The Stein Effect: an Alternative Film-induced Tourism Perspective

International Journal of Tourism Research ◽

10.1002/jtr.1875 ◽

2012 ◽

Vol 15 (6) ◽

pp. 570-582 ◽

Cited By ~ 16

Author(s):

Graham Busby ◽

Rong Huang ◽

Rebecca Jarman

Keyword(s):

Stein Effect

Download Full-text

On the Stein Effect Under Density Power Divergence Loss

Afrika Statistika ◽

10.16929/as/2019.2099.150 ◽

2019 ◽

Vol 14 (3) ◽

pp. 2099-2125

Author(s):

younes Ommane ◽

Idir Ouassou

Keyword(s):

Density Power Divergence ◽

Power Divergence ◽

Stein Effect

Download Full-text

Data-driven Random Fourier Features using Stein Effect

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence ◽

10.24963/ijcai.2017/207 ◽

2017 ◽

Cited By ~ 2

Author(s):

Wei-Cheng Chang ◽

Chun-Liang Li ◽

Yiming Yang ◽

Barnabás Póczos

Keyword(s):

Monte Carlo ◽

Large Scale ◽

Randomized Algorithm ◽

Data Driven ◽

Equal Weight ◽

Quasi Monte Carlo ◽

Empirical Risk ◽

Kernel Approximation ◽

Random Fourier Features ◽

Stein Effect

Large-scale kernel approximation is an important problem in machine learning research. Approaches using random Fourier features have become increasingly popular \cite{Rahimi_NIPS_07}, where kernel approximation is treated as empirical mean estimation via Monte Carlo (MC) or Quasi-Monte Carlo (QMC) integration \cite{Yang_ICML_14}. A limitation of the current approaches is that all the features receive an equal weight summing to 1. In this paper, we propose a novel shrinkage estimator from "Stein effect", which provides a data-driven weighting strategy for random features and enjoys theoretical justifications in terms of lowering the empirical risk. We further present an efficient randomized algorithm for large-scale applications of the proposed method. Our empirical results on six benchmark data sets demonstrate the advantageous performance of this approach over representative baselines in both kernel approximation and supervised learning tasks.

Download Full-text