Abstract:Multiple kernel subspace clustering (MKSC), as an important extension for handling multi-view non-linear subspace data, has shown notable success in a wide variety of machine learning tasks. The key objective of MKSC is to build a flexible and appropriate graph for clustering from the kernel space. However, existing MKSC methods apply a mechanism utilizing the kernel trick to the traditional self-expressive principle, where the similarity graphs are built on the respective high-dimensional (or even infinite) reproducing kernel Hilbert space (RKHS). Regarding this strategy, we argue that the original high-dimensional spaces usually include noise and unreliable similarity measures and, therefore, output a low-quality graph matrix, which degrades clustering performance. In this paper, inspired by projective clustering, we propose the utilization of a complementary similarity graph by fusing the multiple kernel graphs constructed in the low-dimensional partition space, termed projective multiple kernel subspace clustering (PMKSC). By incorporating intrinsic structures with multi-view data, PMKSC alleviates the noise and redundancy in the original kernel space and obtains high-quality similarity to uncover the underlying clustering structures. Furthermore, we design a three-step alternate algorithm with proven convergence to solve the proposed optimization problem. The experimental results on ten multiple kernel benchmark datasets validate the effectiveness of our proposed PMKSC, compared to the state-of-the-art multiple kernel and kernel subspace clustering methods, by a large margin. Our code is available at https://github.com/MengjingSun/PMKSC-code.

Matrix-Based Kernel Subspace Methods

Preservation of Local Linearity by Neighborhood Subspace Scaling for Solving the Pre-Image Problem.

Matrix-based Kernel Principal Component Analysis for Large-Scale Data Set.

Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices

Learning the kernel matrix by resampling.

Projective Multiple Kernel Subspace Clustering.

Class Mean Vector Component and Discriminant Analysis

Directional Principal Component Analysis for Image Matrix

Kernel Multivariate Analysis Framework for Supervised Subspace Learning: A Tutorial on Linear and Kernel Multivariate Methods

Low Rank Matrix Differential Equations As Deep Kernel Principal Component Analysis

Positive Definite Kernels, Algorithms, Frames, and Approximations

Kernelization of Matrix Updates, when and How?

Parallel Approximate Matrix Factorization For Kernel Methods

A General Kernelization Framework for Learning Algorithms Based on Kernel PCA.

Kernel Spectral Curvature Clustering (KSCC)

Visual Clustering Based on Kernel Sparse Representation on Grassmann Manifolds

Computation and Store Space Constrained-Based Sparse Kernel Data Analysis

Sparse Data-Dependent Kernel Principal Component Analysis Based on Least Squares Support Vector Machine for Feature Extraction and Recognition.

Data-Driven Linear Complexity Low-Rank Approximation of General Kernel Matrices: A Geometric Approach

A hybrid stochastic interpolation and compression method for kernel matrices

Combinatorial Kernel Matrix Model Selection Using Feature Distances