Abstract:Kernel methods have been successfully applied in various applications. To succeed in these applications, it is crucial to learn a good kernel representation, whose objective is to reveal the data similarity precisely. In this paper, we address the problem of multiple kernel learning (MKL), searching for the optimal kernel combination weights through maximizing a generalized performance measure. Most MKL methods employ the L(1)-norm simplex constraints on the kernel combination weights, which therefore involve a sparse but non-smooth solution for the kernel weights. Despite the success of their efficiency, they tend to discard informative complementary or orthogonal base kernels and yield degenerated generalization performance. Alternatively, imposing the L(p)-norm (p > 1) constraint on the kernel weights will keep all the information in the base kernels. This leads to non-sparse solutions and brings the risk of being sensitive to noise and incorporating redundant information. To tackle these problems, we propose a generalized MKL (GMKL) model by introducing an elastic-net-type constraint on the kernel weights. More specifically, it is an MKL model with a constraint on a linear combination of the L(1)-norm and the squared L(2)-norm on the kernel weights to seek the optimal kernel combination weights. Therefore, previous MKL problems based on the L(1)-norm or the L(2)-norm constraints can be regarded as special cases. Furthermore, our GMKL enjoys the favorable sparsity property on the solution and also facilitates the grouping effect. Moreover, the optimization of our GMKL is a convex optimization problem, where a local solution is the global optimal solution. We further derive a level method to efficiently solve the optimization problem. A series of experiments on both synthetic and real-world datasets have been conducted to show the effectiveness and efficiency of our GMKL.

Parallel Multiple Kernel Learning: a Hybrid Alternating Direction Method of Multipliers

Generalized Convexity-Based Inexact Projection Method for Multiple Kernel Learning

Parallel multiple kernel learning

A Primal Method for Multiple Kernel Learning

Triply Stochastic Gradients on Multiple Kernel Learning

Two-stage multiple kernel learning with multiclass kernel polarization.

Coupled Multiple Kernel Learning for Supervised Classification.

Efficient Multiple Kernel Learning Algorithms Using Low-Rank Representation.

CLASS-PAIR-GUIDED MULTIPLE KERNEL LEARNING OF INTEGRATING HETEROGENEOUS FEATURES FOR CLASSIFICATION

An Improved Domain Multiple Kernel Support Vector Machine

Multi-kernel Classification Machine with Reduced Complexity

Multiple graph kernel learning based on GMDH-type neural network

Fast randommultiple kernel learning for classification

Multiple Kernel Learning with Hybrid Kernel Alignment Maximization.

Unsupervised Multiple Kernel Learning

Absent Multiple Kernel Learning Algorithms.

Simultaneous Global and Local Graph Structure Preserving for Multiple Kernel Clustering

Efficient sparse generalized multiple kernel learning

A Novel Multiple Nyström-approximating Kernel Discriminant Analysis

Multiple Kernel Subspace Learning for Clustering and Classification

Ideal Kernel-Based Multiple Kernel Learning for Spectral-Spatial Classification of Hyperspectral Image.