High-Dimensional Analysis of Bootstrap Ensemble Classifiers
This work addresses the problem of understanding and improving bootstrap ensemble methods for researchers and practitioners in machine learning, though it is incremental as it builds on existing techniques.
The paper tackles the theoretical analysis of bootstrap ensemble classifiers, specifically for Least Square Support Vector Machines in high-dimensional settings, and proposes strategies to optimize performance, with empirical validation on synthetic and real-world datasets.
Bootstrap methods have long been a cornerstone of ensemble learning in machine learning. This paper presents a theoretical analysis of bootstrap techniques applied to the Least Square Support Vector Machine (LSSVM) ensemble in the context of large and growing sample sizes and feature dimensionalities. Leveraging tools from Random Matrix Theory, we investigate the performance of this classifier that aggregates decision functions from multiple weak classifiers, each trained on different subsets of the data. We provide insights into the use of bootstrap methods in high-dimensional settings, enhancing our understanding of their impact. Based on these findings, we propose strategies to select the number of subsets and the regularization parameter that maximize the performance of the LSSVM. Empirical experiments on synthetic and real-world datasets validate our theoretical results.