Model Tree

Model trees are a class of machine learning models that combine the interpretability of decision trees with the flexibility of more complex models, serving as both predictive tools and insightful surrogate models for interpreting "black box" systems. Current research focuses on applying model trees to diverse tasks, including diagnosing neural network failures by analyzing loss landscapes, reconstructing the lineage of related models (Model Tree Heritage Recovery), and improving the interpretability of complex models through model distillation. This work is significant for enhancing the transparency and explainability of machine learning, leading to more trustworthy and reliable applications across various domains, including data stream analysis and model interpretation.