Shrub Ensembles for Online Classification

TL;DR

This paper introduces shrub ensembles, a memory-efficient online classification method that trains small decision trees and learns ensemble weights with stochastic proximal gradient descent, outperforming many existing methods in resource-constrained environments.

Contribution

The paper proposes a novel shrub ensemble algorithm that improves memory efficiency and performance in online classification tasks, with theoretical analysis and extensive empirical validation.

Findings

Outperforms 8 state-of-the-art methods in 7 of 12 datasets.

Retains high accuracy with limited memory resources.

Provides a better accuracy-memory trade-off in resource-constrained settings.

Abstract

Online learning algorithms have become a ubiquitous tool in the machine learning toolbox and are frequently used in small, resource-constraint environments. Among the most successful online learning methods are Decision Tree (DT) ensembles. DT ensembles provide excellent performance while adapting to changes in the data, but they are not resource efficient. Incremental tree learners keep adding new nodes to the tree but never remove old ones increasing the memory consumption over time. Gradient-based tree learning, on the other hand, requires the computation of gradients over the entire tree which is costly for even moderately sized trees. In this paper, we propose a novel memory-efficient online classification ensemble called shrub ensembles for resource-constraint systems. Our algorithm trains small to medium-sized decision trees on small windows and uses stochastic proximal gradient descent to learn the ensemble weights of these `shrubs'. We provide a theoretical analysis of our algorithm and include an extensive discussion on the behavior of our approach in the online setting. In a series of 2~959 experiments on 12 different datasets, we compare our method against 8 state-of-the-art methods. Our Shrub Ensembles retain an excellent performance even when only little memory is available. We show that SE offers a better accuracy-memory trade-off in 7 of 12 cases, while having a statistically significant better performance than most other methods. Our implementation is available under https://github.com/sbuschjaeger/se-online .