Consistency Analysis of Nearest Subspace Classifier

TL;DR

This paper analyzes the consistency and effectiveness of the Nearest Subspace Classifier (NSS), demonstrating its strong theoretical guarantees and practical efficiency for large-scale data classification.

Contribution

It proves NSS is strongly consistent under certain conditions and evaluates its performance, highlighting its efficiency and effectiveness compared to other linear classifiers.

Findings

NSS is strongly consistent under specific assumptions.

NSS achieves effective classification results on various datasets.

NSS is computationally efficient, especially for large-scale data.

Abstract

The Nearest subspace classifier (NSS) finds an estimation of the underlying subspace within each class and assigns data points to the class that corresponds to its nearest subspace. This paper mainly studies how well NSS can be generalized to new samples. It is proved that NSS is strongly consistent under certain assumptions. For completeness, NSS is evaluated through experiments on various simulated and real data sets, in comparison with some other linear model based classifiers. It is also shown that NSS can obtain effective classification results and is very efficient, especially for large scale data sets.