Fast Meta-Learning for Adaptive Hierarchical Classifier Design

TL;DR

This paper introduces a fast, hierarchical meta-learning approach for multiclass classification that adaptively merges classes based on empirical Bayes error estimates, resulting in quicker learning with competitive accuracy.

Contribution

It presents a novel splitting criterion using Henze-Penrose bounds and MST-based empirical Bayes error estimates for efficient hierarchical classifier design.

Findings

Faster learning compared to existing methods.

Achieves competitive classification accuracy.

Effective hierarchical structure construction.

Abstract

We propose a new splitting criterion for a meta-learning approach to multiclass classifier design that adaptively merges the classes into a tree-structured hierarchy of increasingly difficult binary classification problems. The classification tree is constructed from empirical estimates of the Henze-Penrose bounds on the pairwise Bayes misclassification rates that rank the binary subproblems in terms of difficulty of classification. The proposed empirical estimates of the Bayes error rate are computed from the minimal spanning tree (MST) of the samples from each pair of classes. Moreover, a meta-learning technique is presented for quantifying the one-vs-rest Bayes error rate for each individual class from a single MST on the entire dataset. Extensive simulations on benchmark datasets show that the proposed hierarchical method can often be learned much faster than competing methods, while achieving competitive accuracy.