Learning near-optimal decisions: from SAA to robust optimization
Published in In Preparation, 2022
Recommended citation: Cristian, R., Perakis, G. (2022). Learning Robust Decisions for Contextual Stochastic Optimization Problems Directly from Data
We propose a robust method in solving the end-to-end problems. We propose a novel approach that directly recommends robust, near-optimal decisions based on data. Our proposed approach is based on an optimization problem we introduce that encompasses a feature-based Sample Average Approximation (SAA) approach and robust optimization as special cases.