From Sequential to Recursive: Enhancing Decision-Focused Learning with Bidirectional Feedback

TL;DR

This paper introduces recursive decision-focused learning (R-DFL), a novel framework that incorporates bidirectional feedback between prediction and optimization, improving decision quality in complex decision-making scenarios.

Contribution

The paper proposes R-DFL, extending existing DFL by enabling bidirectional feedback and developing two differentiation methods with comparable accuracy and improved efficiency.

Findings

R-DFL outperforms sequential DFL in decision quality.

Implicit differentiation offers better computational efficiency.

R-DFL demonstrates robustness across synthetic and real-world datasets.

Abstract

Decision-focused learning (DFL) has emerged as a powerful end-to-end alternative to conventional predict-then-optimize (PTO) pipelines by directly optimizing predictive models through downstream decision losses. Existing DFL frameworks are limited by their strictly sequential structure, referred to as sequential DFL (S-DFL). However, S-DFL fails to capture the bidirectional feedback between prediction and optimization in complex interaction scenarios. In view of this, we first time propose recursive decision-focused learning (R-DFL), a novel framework that introduces bidirectional feedback between downstream optimization and upstream prediction. We further extend two distinct differentiation methods: explicit unrolling via automatic differentiation and implicit differentiation based on fixed-point methods, to facilitate efficient gradient propagation in R-DFL. We rigorously prove that both methods achieve comparable gradient accuracy, with the implicit method offering superior computational efficiency. Extensive experiments on both synthetic and real-world datasets, including the newsvendor problem and the bipartite matching problem, demonstrate that R-DFL not only substantially enhances the final decision quality over sequential baselines but also exhibits robust adaptability across diverse scenarios in closed-loop decision-making problems.