Task-based End-to-end Model Learning in Stochastic Optimization

TLDR

Machine learning models are increasingly embedded in larger processes, yet training objectives often diverge from the ultimate evaluation criteria. The study proposes an end‑to‑end method for learning probabilistic models that directly optimizes the task‑based objective in stochastic programming. The method is evaluated on three tasks: inventory stock, electrical grid scheduling, and energy storage arbitrage. The approach outperforms traditional modeling and black‑box policy optimization in all three applications.

Abstract

With the increasing popularity of machine learning techniques, it has become common to see prediction algorithms operating within some larger process. However, the criteria by which we train these algorithms often differ from the ultimate criteria on which we evaluate them. This paper proposes an end-to-end approach for learning probabilistic machine learning models in a manner that directly captures the ultimate task-based objective for which they will be used, within the context of stochastic programming. We present three experimental evaluations of the proposed approach: a classical inventory stock problem, a real-world electrical grid scheduling task, and a real-world energy storage arbitrage task. We show that the proposed approach can outperform both traditional modeling and purely black-box policy optimization approaches in these applications.