Goal-driven Bayesian Optimal Experimental Design for Robust Decision-Making Under Model Uncertainty
2026-05-25 • Machine Learning
Machine Learning
AI summaryⓘ
The authors explain that traditional methods for choosing experiments focus on reducing uncertainty about model parameters, but this doesn’t always help with making better decisions. They introduce GoBOED, a method that picks experiments specifically to improve decision-making outcomes. By combining advanced math tools, GoBOED focuses only on the parts of the model that matter for decisions. Their tests show GoBOED works better in real tasks like finding sources, managing disease outbreaks, and controlling medicine doses, often giving more flexible experiment options than older methods.
Bayesian optimal experimental designinformation gainvariational posteriordecision-making objectivegradient-based optimizationsource localizationepidemic managementpharmacokineticsconvex optimizationamortized inference
Authors
Jinwoo Go, Xiaoning Qian, Byung-Jun Yoon
Abstract
Bayesian optimal experimental design (BOED) selects experiments to maximize information gain about model parameters. However, in decision-critical settings, reducing parameter uncertainty does not necessarily improve downstream decisions, as only specific parameter directions relevant to the objective truly matter. We propose GoBOED, a goal-driven BOED framework that directly optimizes experimental designs for a specified decision-making objective. GoBOED combines an amortized variational posterior surrogate with a differentiable convex decision layer, enabling gradient-based design optimization that is fully decision-focused. We theoretically show that GoBOED gradients are insensitive to parameter directions irrelevant to the decision objective, providing a formal justification for why goal-driven design achieves equivalent decision quality over a wider set of experimental designs than information-gain maximization. Empirically, across source localization, epidemic management, and pharmacokinetic control, GoBOED identifies designs that better align with downstream decision objectives and reveals that near-optimal design windows are substantially wider than those predicted by goal-agnostic BOED approaches.