Uncertainty-Guided Edge Learning for Deep Image Regression in Remote Sensing

2026-05-06

AI Summary

This work addresses the challenge of efficiently evaluating predictive uncertainty for deep image regression models on resource-constrained remote sensing satellite edge devices, where limited computational capacity hinders effective model updating with unlabeled data. To overcome this limitation, we propose Uncertainty-Guided Edge Learning (UGEL), a novel algorithm that leverages lightweight deep Beta regression to estimate prediction uncertainty under flexible distributional assumptions within a single forward pass. UGEL integrates an active learning strategy to prioritize the selection of high-value samples for labeling and model refinement. Experimental results demonstrate that UGEL significantly accelerates the convergence of onboard regression models compared to existing approaches, achieving superior edge learning efficiency in remote sensing image regression tasks. The implementation code and models are publicly released.

Abstract

Edge learning refers to training machine learning models deployed on edge platforms, typically using new data accumulated onboard. The computational limitations on edge devices affect not only model optimisation, but also calculation of the predictive uncertainty of the current model on the unlabelled data, which is vital for informing model updating. In this paper, we investigate edge learning in the context of performing deep image regression on a remote sensing satellite, where a deep network is executed by an onboard computer to regress a scalar $y$ from an input image, e.g., $y$ is the percentage of pixels indicating cloud coverage or land use. We propose an uncertainty-guided edge learning (UGEL) algorithm that can accurately prioritise the data to speed up training convergence of the on-board regression model. Underpinning UGEL is the calculation of predictive uncertainty based on deep beta regression, where a deep network is used to estimate the parameters of a beta distribution for which the target $y$ for an input image has a high likelihood. Compared to established methods for uncertainty estimation that are either too costly on edge devices (e.g., require many forward passes per sample) or make strict assumptions on the predictive distribution (e.g., Gaussian), deep beta regression is computable in a single forward pass and allows more general predictive distributions. Results show that UGEL delivers faster-converging edge learning than active or semi-supervised learning. Code and models are publicly available at https://github.com/anh-vunguyen/UGEL.

Problem

• edge learning
• predictive uncertainty
• deep image regression
• remote sensing
• onboard training

Innovation

• uncertainty-guided edge learning
• deep beta regression
• predictive uncertainty
• remote sensing
• onboard model updating