Project: Interpretability and explainability methods for tree-based ensemble models: theoretical developments and economic applications.
This project explores the intersection of advanced Machine Learning and decision-making transparency. Although tree-based ensemble models (such as Random Forest and XGBoost) provide enhanced predictive performance, their "black-box" nature often hinders their adoption in highly regulated sectors. This research aims to develop and refine Explainable AI (XAI) methods to make these models interpretable without compromising accuracy. The theoretical framework investigates novel feature attribution metrics, while the applied component validates these tools using economic datasets, addressing complex issues such as credit scoring, financial market forecasting, and public policy analysis.
Duration: 12 months
Lead Researcher: Prof. Massimo Aria
Requirements: PhD in Economics or Statistics
Start Deadline: 01 July 2026
J-18808-Ljbffr

📌 Researcher - Advanced Machine Learning and Decision Making Transparency (Italia)
🏢 Altro
📍 Italia