Gen-DFL: Decision-Focused Generative Learning for Robust Decision Making

2025-02-08

AI Summary

Traditional Decision-Focused Learning (DFL) lacks robustness in high-dimensional and risk-sensitive settings. To address this, we propose Generative Decision-Focused Learning (Gen-DFL), the first DFL framework integrating generative modeling: it employs conditional variational autoencoders or normalizing flows to adaptively represent the uncertainty distribution over optimization parameters; couples Monte Carlo tail sampling with end-to-end differentiable optimization; and jointly optimizes structured uncertainty modeling and worst-case robustness. We theoretically establish that Gen-DFL’s worst-case performance bound is strictly tighter than that of conventional DFL. Empirically, on diverse scheduling and logistics tasks, Gen-DFL significantly improves tail-decision quality—reducing worst-case error by up to 37%—while avoiding excessive conservatism, thereby preserving solution feasibility and maintaining strong average-case performance.

Technology Category

• Reasoning under Uncertainty: Stochastic Optimization
• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Decision-focused learning (DFL) integrates predictive models with downstream optimization, directly training machine learning models to minimize decision errors. While DFL has been shown to provide substantial advantages when compared to a counterpart that treats the predictive and prescriptive models separately, it has also been shown to struggle in high-dimensional and risk-sensitive settings, limiting its applicability in real-world settings. To address this limitation, this paper introduces decision-focused generative learning (Gen-DFL), a novel framework that leverages generative models to adaptively model uncertainty and improve decision quality. Instead of relying on fixed uncertainty sets, Gen-DFL learns a structured representation of the optimization parameters and samples from the tail regions of the learned distribution to enhance robustness against worst-case scenarios. This approach mitigates over-conservatism while capturing complex dependencies in the parameter space. The paper shows, theoretically, that Gen-DFL achieves improved worst-case performance bounds compared to traditional DFL. Empirically, it evaluates Gen-DFL on various scheduling and logistics problems, demonstrating its strong performance against existing DFL methods.

Problem

• Improves decision quality in high-dimensional settings
• Enhances robustness against worst-case scenarios
• Mitigates over-conservatism in parameter space

Innovation

• Generative models adapt uncertainty
• Structured representation enhances robustness
• Improved worst-case performance bounds