litlas
ログインアカウント作成

A Kernel Statistical Test of Independence

著者: Arthur Gretton, Kenji Fukumizu, Choon Hui Teo, Le Song, Bernhard Schölkopf, Alexander J. Smola - Conference on Neural Information Processing Systems, NeurIPS 2007 被引用: 1,068

Although kernel measures of independence have been widely applied in machine learning (notably in kernel ICA), there is as yet no method to determine whether they have detected statistically significant dependence.We provide a novel test of the independence hypothesis for one particular kernel independence measure, the Hilbert-Schmidt independence criterion (HSIC).The resulting test costs O(m 2 ), where m is the sample size.We demonstrate that this test outperforms established contingency table and functional correlation-based tests, and that this advantage is greater for multivariate data.Finally, we show the HSIC test also applies to text (and to structured data more generally), for which no other independence test presently exists.

On measures ofdependenceOn measures of dependenceA Quadratic Measure ofDeviation of…A Quadratic Measure of Deviation of Two-Dimensional Density Estimates and A Test of IndependenceApproximation Theoremsof Mathematical…Approximation Theorems of Mathematical Statistics.Continuous UnivariateDistributions.Continuous Univariate Distributions.A Probabilistic Theoryof Pattern RecognitionA Probabilistic Theory of Pattern Recognition10.1162/15324430276018525210.1162/15324430276018525210.1162/15324430376896608510.1162/153244303768966085Dimensionality Reductionfor Supervised Learning…Dimensionality Reduction for Supervised Learning with Reproducing Kernel Hilbert SpacesKernel Methods inComputational BiologyKernel Methods in Computational BiologyMeasuring StatisticalDependence with…Measuring Statistical Dependence with Hilbert-Schmidt NormsKernel Methods forMeasuring IndependenceKernel Methods for Measuring IndependenceSupervised FeatureSelection via Dependenc…Supervised Feature Selection via Dependence Estimationopenalex_id:w2963535485openalex_id:w2963535485On the Identifiabilityof the Post-Nonlinear…On the Identifiability of the Post-Nonlinear Causal ModelIdentifiability ofCausal Graphs using…Identifiability of Causal Graphs using Functional ModelsKernel Bayes' RuleKernel Bayes' RuleEquivalence ofdistance-based and…Equivalence of distance-based and RKHS-based statistics in hypothesis testingCausal discovery withcontinuous additive…Causal discovery with continuous additive noise modelsKernel Bayes' rule:Bayesian inference with…Kernel Bayes' rule: Bayesian inference with positive definite kernelsCausal Inference on TimeSeries using Restricted…Causal Inference on Time Series using Restricted Structural Equation ModelsMeasuring DependencePowerfully and EquitablyMeasuring Dependence Powerfully and EquitablyDistance Metrics forMeasuring Joint…Distance Metrics for Measuring Joint Dependence with Application to Causal InferenceGaussian Processes andKernel Methods: A Revie…Gaussian Processes and Kernel Methods: A Review on Connections and EquivalencesKernel DistributionEmbeddings: Universal…Kernel Distribution Embeddings: Universal Kernels, Characteristic Kernels and Kernel Metrics on DistributionsA Kernel StatisticalTest of IndependenceA Kernel Statistical Test of Independence過去の参考文献中心の論文この論文を引用する論文古い新しい

ノードをクリックするとフォーカスを固定、空白をクリックすると本論文に戻ります。ホバーで一時的にプレビューできます。各ノードのページはタイトルから開けます。