A Proxy Consistency Loss for Grounded Fusion of Earth Observation and Location Encoders

2026-04-20

AI Summary

This work addresses the challenge of limited high-quality labels in Earth observation tasks by proposing a trainable positional encoder that flexibly integrates abundant yet correlated proxy geospatial data. To enhance model generalization under label scarcity, the authors introduce a Proxy Consistency Loss (PCL) to regularize the learning process. Unlike approaches relying on frozen geographic embeddings or naive input concatenation, the proposed method adaptively leverages auxiliary data through end-to-end training. Experiments on air quality estimation and poverty mapping demonstrate substantial improvements over existing baselines, with consistently stronger predictive performance both within and beyond the observed regions, highlighting its robustness and generalizability in data-scarce settings.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Supervised learning with Earth observation inputs is often limited by the sparsity of high-quality labeled or in-situ measured data to use as training labels. With the abundance of geographic data products, in many cases there are variables correlated with - but different from - the variable of interest that can be leveraged. We integrate such proxy variables within a geographic prior via a trainable location encoder and introduce a proxy consistency loss (PCL) formulation to imbue proxy data into the location encoder. The first key insight behind our approach is to use the location encoder as an agile and flexible way to learn from abundantly available proxy data which can be sampled independently of training label availability. Our second key insight is that we will need to regularize the location encoder appropriately to achieve performance and robustness with limited labeled data. Our experiments on air quality prediction and poverty mapping show that integrating proxy data implicitly through the location encoder outperforms using both as input to an observation encoder and fusion strategies that use frozen, pretrained location embeddings as a geographic prior. Superior performance for in-sample prediction shows that the PCL can incorporate rich information from the proxies, and superior out-of-sample prediction shows that the learned latent embeddings help generalize to areas without training labels.

Problem

• Earth observation
• proxy variables
• supervised learning
• data sparsity
• geographic prior

Innovation

• Proxy Consistency Loss
• Location Encoder
• Earth Observation
• Geographic Prior
• Data-scarce Learning