Learning With Selected Features

doi:10.1109/TCYB.2020.2987810

Learning With Selected Features

2022-04-01

3

OA

AI

DOI：10.1109/TCYB.2020.2987810

原文链接

查看原文

摘要

En 中文

The coming big data era brings data of unprecedented size and launches an innovation of learning algorithms in statistical and machine-learning communities. The classical kernel-based regularized least-squares (RLS) algorithm is excluded in the innovation, due to its computational and storage bottlenecks. This article presents a scalable algorithm based on subsampling, called learning with selected features (LSF), to reduce the computational burden of RLS. Almost the optimal learning rate together with a sufficient condition on selecting kernels and centers to guarantee the optimality is derived. Our theoretical assertions are verified by numerical experiments, including toy simulations, UCI standard data experiments, and a real-world massive data application. The studies in this article show that LSF can reduce the computational burden of RLS without sacrificing its generalization ability very much.

Keyword:

Kernel

Computational complexity

Training

Technological innovation

Machine learning

Standards

Cybernetics

Learning theory

regularized least squares (RLS)

selected features

subsampling

uniqueness set

Learning With Selected Features

摘要

AI总结

期刊

机构