When does Diversity Help Generalization in Classification Ensembles?

TL;DR

This paper investigates how diversity affects the generalization of classification ensembles, revealing that increasing diversity improves generalization only within specific ranges and proposing pruning methods to optimize ensemble performance.

Contribution

It introduces a novel diversity measurement based on error decomposition and establishes a theoretical relationship with generalization error, along with practical pruning methods.

Findings

Diversity improves generalization only in certain ranges.

The proposed pruning methods effectively balance diversity and ensemble size.

Empirical results support the theoretical relationship between diversity and generalization.

Abstract

Ensembles, as a widely used and effective technique in the machine learning community, succeed within a key element -- "diversity." The relationship between diversity and generalization, unfortunately, is not entirely understood and remains an open research issue. To reveal the effect of diversity on the generalization of classification ensembles, we investigate three issues on diversity, i.e., the measurement of diversity, the relationship between the proposed diversity and the generalization error, and the utilization of this relationship for ensemble pruning. In the diversity measurement, we measure diversity by error decomposition inspired by regression ensembles, which decomposes the error of classification ensembles into accuracy and diversity. Then we formulate the relationship between the measured diversity and ensemble performance through the theorem of margin and generalization and observe that the generalization error is reduced effectively only when the measured diversity is increased in a few specific ranges, while in other ranges larger diversity is less beneficial to increasing the generalization of an ensemble. Besides, we propose two pruning methods based on diversity management to utilize this relationship, which could increase diversity appropriately and shrink the size of the ensemble without much-decreasing performance. Empirical results validate the reasonableness of the proposed relationship between diversity and ensemble generalization error and the effectiveness of the proposed pruning methods.