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

TL;DR

This paper introduces UGEL, an uncertainty-guided edge learning algorithm for deep image regression on remote sensing satellites, enabling faster training convergence by efficiently estimating predictive uncertainty with a single forward pass.

Contribution

The paper proposes a novel uncertainty estimation method using deep beta regression that is computationally efficient for edge devices and improves training speed in remote sensing image regression.

Findings

UGEL achieves faster convergence than active or semi-supervised learning.

Deep beta regression provides accurate uncertainty estimates with a single forward pass.

The method is suitable for resource-constrained edge platforms in remote sensing applications.

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.