Supervised Learning via Ensembles of Diverse Functional Representations: the Functional Voting Classifier

TL;DR

The paper addresses classification of high-dimensional functional data by leveraging ensemble learning. It introduces the Functional Voting Classifier (FVC), which trains base learners on multiple, diverse functional representations (via different B-spline orders) and combines their predictions through majority voting. Empirical results on ten real-world datasets show that FVC often yields higher accuracy than any single model, with diversity between representations correlating with performance gains. The approach provides a practical framework for exploiting functional representations to improve robustness and generalization in functional data classification, while noting trade-offs in interpretability and computation.

Abstract

Many conventional statistical and machine learning methods face challenges when applied directly to high dimensional temporal observations. In recent decades, Functional Data Analysis (FDA) has gained widespread popularity as a framework for modeling and analyzing data that are, by their nature, functions in the domain of time. Although supervised classification has been extensively explored in recent decades within the FDA literature, ensemble learning of functional classifiers has only recently emerged as a topic of significant interest. Thus, the latter subject presents unexplored facets and challenges from various statistical perspectives. The focal point of this paper lies in the realm of ensemble learning for functional data and aims to show how different functional data representations can be used to train ensemble members and how base model predictions can be combined through majority voting. The so-called Functional Voting Classifier (FVC) is proposed to demonstrate how different functional representations leading to augmented diversity can increase predictive accuracy. Many real-world datasets from several domains are used to display that the FVC can significantly enhance performance compared to individual models. The framework presented provides a foundation for voting ensembles with functional data and can stimulate a highly encouraging line of research in the FDA context.

Figures (20)

• Figure 1: Diversity on the FordA dataset.
• Figure 2: Diversity on the ECG5000 dataset.
• Figure 3: Diversity on the ChlorineConcentration dataset.
• Figure 4: Diversity on the Computers dataset.
• Figure 5: Diversity on the GunPoint dataset.