Foundation model embeddings improve urban livability predictions
Applying foundation model embeddings towards urban livability evaluation
Machine LearningComputers and Society
Summary
Measuring how livable a city is can be hard in places without much data. The authors look at special AI tools called foundation model embeddings that analyze satellite images to find clues about cities’ livability. They found a way to pick the best types of satellite data to make these predictions more accurate. This helps people understand which city features matter most for making places better to live, especially where traditional data is missing.
What this means in practice
- •For urban planning teams: Use foundation model embeddings from satellite images to improve predictions of livability factors in cities lacking detailed surveys.
- •For nonprofit aid organizations: Identify priority regions for policy interventions by ranking urban areas based on predicted socioeconomic conditions derived from satellite data embeddings.
- •For real estate analysts: Assess neighborhood livability using satellite-driven AI features to better inform property investment decisions when ground data is unavailable.$Commercial implications: Enables development of AI-based real estate tools to estimate livability metrics for investment firms without access to local surveys.
Authors
Ayush Khot, Wen Zhou, Shaowen Wang
Abstract
While accurate measurement of socioeconomic indicators remains challenging in data-scarce regions, which limits policy interventions and resource allocation, high-resolution geospatial data is widely available and can contain information on various livability statistics. We investigate which physical features are encoded within foundation model embeddings, such as AlphaEarth, AnySat, and TerraMind, and provide a systematic framework for identifying the most predictive geospatial indicators. By analyzing how different types of geospatial data influence urban livability predictions, our approach enables researchers to prioritize the most informative features for their specific applications. Additionally, we demonstrate how to leverage foundation model embeddings to enhance prediction performance for these outcomes. This work contributes a principled methodology for extracting actionable information from satellite imagery while accounting for complex spatial dependencies, with applications in predicting urban livability in regions with limited observation data.