OPO: Making Decision-Focused Data Acquisition Decisions

2025-04-21

AI Summary

In contextual stochastic optimization, acquiring context variables is costly and budget-constrained, limiting effective decision-making. Method: This paper proposes an end-to-end “Optimize–Predict–Optimize” (OPO) framework—the first to incorporate differentiable optimization into the data acquisition stage, directly optimizing for downstream decision quality rather than heuristic or coverage-based surrogate objectives. Leveraging decision-focused learning, linear surrogate modeling, and end-to-end gradient backpropagation, OPO jointly optimizes the entire acquisition–prediction–decision pipeline. Results: Evaluated on UAV reconnaissance path planning, OPO significantly improves shortest-path decision accuracy and robustness under limited acquisition budgets, outperforming baselines such as random search. Its core contribution is establishing a decision-quality-driven, differentiable data acquisition paradigm that overcomes the context variable selection bottleneck under high-cost constraints.

Technology Category

• Search and Optimization: Learning to Search
• Reasoning under Uncertainty: Stochastic Optimization
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

We propose a model for making data acquisition decisions for variables in contextual stochastic optimisation problems. Data acquisition decisions are typically treated as separate and fixed. We explore problem settings in which the acquisition of contextual variables is costly and consequently constrained. The data acquisition problem is often solved heuristically for proxy objectives such as coverage. The more intuitive objective is the downstream decision quality as a result of data acquisition decisions. The whole pipeline can be characterised as an optimise-then-predict-then-optimise (OPO) problem. Analogously, much recent research has focused on how to integrate prediction and optimisation (PO) in the form of decision-focused learning. We propose leveraging differentiable optimisation to extend the integration to data acquisition. We solve the data acquisition problem with well-defined constraints by learning a surrogate linear objective function. We demonstrate an application of this model on a shortest path problem for which we first have to set a drone reconnaissance strategy to capture image segments serving as inputs to a model that predicts travel costs. We ablate the problem with a number of training modalities and demonstrate that the differentiable optimisation approach outperforms random search strategies.

Problem

• Optimizing data acquisition for contextual stochastic problems
• Improving decision quality via differentiable optimization
• Cost-constrained drone reconnaissance for path prediction

Innovation

• Differentiable optimization for data acquisition
• Surrogate linear objective function learning
• Decision-focused learning integration extension