Decision-focused Sensing and Forecasting for Adaptive and Rapid Flood Response: An Implicit Learning Approach

TL;DR

This work tackles the challenge of rapid flood response under limited sensing by introducing a decision-focused, end-to-end framework that jointly optimizes in-situ sensor placement, spatio-temporal flood reconstruction, and downstream disaster decisions. It leverages a location-scoring network, a differentiable sensor sampling module under a hard budget with Implicit Maximum Likelihood Estimation (I-MLE) to propagate gradients through discrete sensor configurations, a spatio-temporal reconstruction/forecasting module, and differentiable decision heads for evacuation and aircraft relocation. Key contributions include (i) the I-MLE-based sensor placement mechanism, (ii) task-specific differentiable decision layers using Sinkhorn normalization and overflow penalties, and (iii) end-to-end training that aligns sensing and forecasting with decision regret, demonstrated on the 2019 Offutt AFB flood with improved evacuation cost, relocation cost, and reconstruction accuracy. The results show that decision-aware sensing and forecasting substantially enhance the speed and reliability of flood-response operations, offering a scalable approach to real-time disaster management under budget constraints.

Abstract

Timely and reliable decision-making is vital for flood emergency response, yet it remains severely hindered by limited and imprecise situational awareness due to various budget and data accessibility constraints. Traditional flood management systems often rely on in-situ sensors to calibrate remote sensing-based large-scale flood depth forecasting models, and further take flood depth estimates to optimize flood response decisions. However, these approaches often take fixed, decision task-agnostic strategies to decide where to put in-situ sensors (e.g., maximize overall information gain) and train flood forecasting models (e.g., minimize average forecasting errors), but overlook that systems with the same sensing gain and average forecasting errors may lead to distinct decisions. To address this, we introduce a novel decision-focused framework that strategically selects locations for in-situ sensor placement and optimize spatio-temporal flood forecasting models to optimize downstream flood response decision regrets. Our end-to-end pipeline integrates four components: a contextual scoring network, a differentiable sensor selection module under hard budget constraints, a spatio-temporal flood reconstruction and forecasting model, and a differentiable decision layer tailored to task-specific objectives. Central to our approach is the incorporation of Implicit Maximum Likelihood Estimation (I-MLE) to enable gradient-based learning over discrete sensor configurations, and probabilistic decision heads to enable differentiable approximation to various constrained disaster response tasks.

Figures (5)

• Figure 1: Framework Architecture.
• Figure 2: Study region and hydrological conditions.
• Figure 3: Visual explanations of flood response tasks.
• Figure 4: Training curves visualization.
• Figure 5: Ablation Studies.