Abstract:Kernel methods are attractive in data analysis as they can model nonlinear similarities between observations and provide means to rich representations, both of which are useful for the regression problems in general domains. Despite their popularity, they suffer from two primary inherent drawbacks. One drawback is the positive definiteness requirement of the kernel functions, which greatly restricts their applications to some real data analysis. The other drawback is their poor scalability in massive data scenarios. In this paper, we aim to address these two problems by considering the Nystrom subsampling approach for coefficient-based regularized regression (or Nystrom CRR for short). Nystrom subsampling is an effective approach to reduce the space and time complexity by constructing a low-rank approximation of the original kernel matrix through column subsampling. Coefficient-based regularized regression can provide a simple paradigm for designing indefinite kernel methods. We show that a combination of these two schemes is not only computationally efficient but also statistically consistent with a mini-max optimal rates of convergence. We explicitly determine the lower bound of subsampling level as a function of the sample size such that the mini-max optimal convergence rates can be preserved. From our analysis, the subsampling level plays a role as a trade-off between computational and asymptotic behaviors of Nystrom CRR, and hence is pivotal for algorithmic performances. In order to choose an appropriate subsampling level, we develop an incremental Nystrom CRR for nested subsampling sets. The proposed algorithm can greatly reduce the cost of cross-validation, and even allows to compute the whole solution path through all possible subsampling levels. Based on our empirical studies, the incremental Nystrom CRR can perform effective model selections and achieve the state-of-the-art results on both synthetic and real data sets.

Improving the Modified Nyström Method Using Spectral Shifting

Efficient Algorithms and Error Analysis for the Modified Nystrom Method.

Towards More Efficient SPSD Matrix Approximation and CUR Matrix Decomposition

Making Large-Scale Nyström Approximation Possible

Large-scale Nyström kernel matrix approximation using randomized SVD.

Making the Nyström method highly accurate for low-rank approximations

SPSD Matrix Approximation Vis Column Selection: Theories, Algorithms, and Extensions

Improved Bounds for the Nyström Method with Application to Kernel Classification

A generalized Nystrom method with column sketching for low-rank approximation of nonsymmetric matrices

Improving CUR Matrix Decomposition and the Nyström Approximation Via Adaptive Sampling

A New Spectral Difference Method Using Hierarchical Polynomial Bases for Hyperbolic Conservation Laws

Dynamic Subspace Update with Incremental Nyström Approximation

Fast and Accurate Refined Nyström-Based Kernel SVM

Improved guarantees and a multiple-descent curve for Column Subset Selection and the Nyström method

Fast Implementation of Kernel Simplex Volume Analysis Based on Modified Cholesky Factorization for Endmember Extraction

Randomized low-rank approximation for symmetric indefinite matrices

Spectral Variation Alleviation by Low-Rank Matrix Approximation for Hyperspectral Image Analysis.

Reduced Heteroscedasticity Linear Regression for Nyström Approximation

Approximating Element-Wise Functions of Matrix with Improved Streaming Randomized SVD

Nyström Subsampling Method for Coefficient-Based Regularized Regression

A multilinear Nyström algorithm for low-rank approximation of tensors in Tucker format