Dynamic Meta-Learning for Adaptive XGBoost-Neural Ensembles

Abstract--This paper introduces a novel adaptive ensemble framework that synergistically combines XGBoost and neural networks through sophisticated meta-learning. The proposed method leverages advanced uncertainty quantification techniques and feature importance integration to dynamically orchestrate model selection and combination. Experimental results demonstrate superior predictive performance and enhanced interpretability across diverse datasets, contributing to the development of more intelligent and flexible machine learning systems. In the rapidly evolving landscape of machine learning, the quest for models that can adapt to complex, heterogeneous data while maintaining high predictive accuracy remains a significant challenge. T raditional approaches often rely on either tree-based methods, such as XGBoost [1], known for their effectiveness with tabular data, or neural networks [2], [14], celebrated for their ability to capture intricate patterns [15].