litlas
ログインアカウント作成

Learning Non-Linear Combinations of Kernels

著者: Corinna Cortes, Mehryar Mohri, Afshin Rostamizadeh - Conference on Neural Information Processing Systems, NeurIPS 2009 被引用: 323

This paper studies the general problem of learning kernels based on a polynomial combination of base kernels. We analyze this problem in the case of regression and the kernel ridge regression algorithm. We examine the corresponding learning kernel optimization problem, show how that minimax problem can be reduced to a simpler minimization problem, and prove that the global solution of this problem always lies on the boundary. We give a projection-based gradient descent algo-rithm for solving the optimization problem, shown empirically to converge in few iterations. Finally, we report the results of extensive experiments with this algo-rithm using several publicly available datasets demonstrating the effectiveness of our technique. 1

Support-Vector NetworksSupport-Vector NetworksNonlinear ComponentAnalysis as a Kernel…Nonlinear Component Analysis as a Kernel Eigenvalue ProblemLearning the Kernel withHyperkernelsLearning the Kernel with HyperkernelsKernel Methods forPattern AnalysisKernel Methods for Pattern AnalysisLearning ConvexCombinations of…Learning Convex Combinations of Continuously Parameterized Basic KernelsLearning the KernelFunction via…Learning the Kernel Function via RegularizationA DC-programmingalgorithm for kernel…A DC-programming algorithm for kernel selectionLearning Bounds forSupport Vector Machines…Learning Bounds for Support Vector Machines with Learned KernelsExploring Large FeatureSpaces with Hierarchica…Exploring Large Feature Spaces with Hierarchical Multiple Kernel LearningLearning sequencekernelsLearning sequence kernelsL2 Regularization forLearning KernelsL2 Regularization for Learning KernelsMore generality inefficient multiple…More generality in efficient multiple kernel learningTwo-Stage LearningKernel AlgorithmsTwo-Stage Learning Kernel AlgorithmsGeneralization Boundsfor Learning KernelsGeneralization Bounds for Learning Kernelslp-Norm Multiple KernelLearninglp-Norm Multiple Kernel LearningSPF-GMKL: generalizedmultiple kernel learnin…SPF-GMKL: generalized multiple kernel learning with a million kernelsLocalized algorithms formultiple kernel learningLocalized algorithms for multiple kernel learningLocal Deep KernelLearning for Efficient…Local Deep Kernel Learning for Efficient Non-linear SVM PredictionLocalized MultipleKernel Learning Via…Localized Multiple Kernel Learning Via Sample-Wise Alternating OptimizationAn Approximate Approachto Automatic Kernel…An Approximate Approach to Automatic Kernel SelectionNonlinear Deep KernelLearning for Image…Nonlinear Deep Kernel Learning for Image AnnotationMatrix-RegularizedMultiple Kernel Learnin…Matrix-Regularized Multiple Kernel Learning via (r, p) NormsTwo-Stage Fuzzy MultipleKernel Learning Based o…Two-Stage Fuzzy Multiple Kernel Learning Based on Hilbert-Schmidt Independence CriterionLocalized MultipleKernel Learning With…Localized Multiple Kernel Learning With Dynamical Clustering and Matrix RegularizationLearning Non-LinearCombinations of KernelsLearning Non-Linear Combinations of Kernels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。