A General Purpose Neural Architecture for Geospatial Systems

TL;DR

This paper proposes a general-purpose neural architecture for geospatial data that can handle diverse modalities and tasks, aiming to improve collaboration and performance in geospatial applications.

Contribution

It introduces a neural architecture with a geospatial inductive bias, trained in a self-supervised manner on large unlabelled earth observation data, to unify diverse geospatial tasks.

Findings

Achieves competitive performance with domain-specific models on SDG-related tasks.

Demonstrates versatility across multiple geospatial data modalities.

Lays groundwork for a collaborative, multi-task geospatial AI model.

Abstract

Geospatial Information Systems are used by researchers and Humanitarian Assistance and Disaster Response (HADR) practitioners to support a wide variety of important applications. However, collaboration between these actors is difficult due to the heterogeneous nature of geospatial data modalities (e.g., multi-spectral images of various resolutions, timeseries, weather data) and diversity of tasks (e.g., regression of human activity indicators or detecting forest fires). In this work, we present a roadmap towards the construction of a general-purpose neural architecture (GPNA) with a geospatial inductive bias, pre-trained on large amounts of unlabelled earth observation data in a self-supervised manner. We envision how such a model may facilitate cooperation between members of the community. We show preliminary results on the first step of the roadmap, where we instantiate an architecture that can process a wide variety of geospatial data modalities and demonstrate that it can achieve competitive performance with domain-specific architectures on tasks relating to the U.N.'s Sustainable Development Goals.