Calibrating Attribution Proxies for Reward Allocation in Participatory Weather Sensing

arXiv cs.LG / 5/1/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper addresses a key open problem in participatory weather-sensing IoT networks: how to translate individual sensor data contributions into a reliable monetary/points reward signal rather than only assessing data quality.
It proposes using differentiable AI weather models to compute gradient-based attribution on gridded GFS analysis inputs, leveraging model gradients to estimate each contribution’s value.
The study evaluates the approach across 400+ configurations, focusing on fidelity (how well attribution reflects true utility), calibration, computational cost, and susceptibility to gaming.
Results show gradient attribution can provide near-optimal guidance for sensor placement and enable monotonically faithful payments, but it can be inflated by adversarial inputs, requiring external baseline data for detection.
Overall, the authors argue that gradient attribution is a computationally validated, model-informed signal suitable for reward allocation in participatory weather sensing.

Abstract

Large-scale IoT weather sensing networks require incentive mechanisms to sustain participation, yet determining how much value individual data contributions bring to the network remains an open problem. Existing approaches address data quality but not data valuation; in operational meteorology, adjoint-based methods derive value from the forecast model itself but require full data assimilation infrastructure. We propose to utilise differentiable AI weather models to fill this gap and characterise gradient-based attribution on gridded GFS analysis inputs as a candidate value signal, evaluating fidelity, calibration, cost, and gaming vulnerability across more than 400 configurations. Attribution captures near-optimal sensor placement utility with monotonically faithful payments, but can be inflated by adversarial inputs, with detection requiring external baseline data. These findings establish gradient attribution as a computationally validated signal for model-informed reward allocation in participatory weather sensing.