Risk-aware sensor placement improves prediction accuracy in experiments

Risk-Aware Goal-Oriented Bayesian Optimal Experimental Design

Computational Engineering, Finance, and Science

Summary

Choosing the best measurements in experiments often aims to understand model parameters, but this doesn’t always help with accurate predictions. The authors create a new approach that focuses on making good predictions, especially when rare but important outcomes matter a lot. They combine different ways to measure risk and uncertainty to tailor the design to what matters most. Their method lets computers efficiently find optimal sensor locations without testing every possibility, and they show it works better than traditional choices in a model problem involving how substances spread. This approach helps make smarter decisions about where to collect data in complex systems.

Bayesian optimal experimental designGoal-oriented designPredictive uncertaintyRisk measuresSensor placementAdvection-diffusion equationGradient-based optimizationLinear-Gaussian modelNested quadraturePosterior predictive distribution

Authors

John D. Jakeman, Rebekah White, Bart van Bloemen Waanders, Drew P. Kouri, Alen Alexanderian

Abstract

Traditional Bayesian optimal experimental design (OED) selects measurements that best inform a model's parameters. However, such measurements can be suboptimal for downstream predictions. Goal-oriented OED targets the prediction directly. However, the existing goal-oriented criteria value all reductions in predictive uncertainty equally, with no way to prioritize rare, high-consequence outcomes. In this article, we develop a risk-aware framework that composes risk at three levels, each generalizing an ingredient of classical $I$- and $G$-optimal design: a deviation measure of the posterior predictive uncertainty (generalizing the predictive variance), a risk measure across the prediction domain (interpolating $I$-optimal averaging and $G$-optimal worst-case selection), and a risk measure over datasets (generalizing the expectation). We generate each level from a regret function in the risk quadrangle, so that one triple specifies a practitioner's risk preference. We relax the design to continuous weights on the unit simplex and construct a nested-quadrature estimator that is differentiable in the design variable. This enables solving the optimal design problem with gradient-based methods, avoiding a combinatorial search over candidate designs. For a linear-Gaussian lognormal model and a nonlinear extension, we derive closed-form objectives. These give exact references against which we verify that the estimator converges. We demonstrate this framework for finding optimal sensor placements in an inverse problem governed by an advection-diffusion equation. We find that the risk-aware designs substantially outperform the expected-information-gain baseline, which is statistically indistinguishable from a random allocation.