Transferable Human Mobility Network Reconstruction with neuroGravity

TL;DR

neuroGravity is a physics-informed deep learning model that reconstructs human mobility networks from limited data, transfers across cities, and reveals socioeconomic factors influencing mobility patterns.

Contribution

the paper introduces neuroGravity, a novel model that reconstructs mobility flows from minimal data and predicts transferability based on socioeconomic segregation.

Findings

neuroGravity accurately reconstructs mobility flows using only urban facility and population data.

mobility network transferability is strongly linked to socioeconomic segregation levels.

the model can generate mobility proxies for over 1,200 cities worldwide.

Abstract

Accurate modeling of human mobility is critical for tackling urban planning and public health challenges. In undeveloped regions, the absence of comprehensive travel surveys necessitates reconstructing mobility networks from publicly available data. Here we develop neuroGravity, a physics-informed deep learning model that reliably reconstructs mobility flows from limited observations and transfers to unobserved cities. Using only urban facility and population distributions, we find that neuroGravity's regional representations strongly correlate with socioeconomic and livability status, offering scalable proxies for costly surveys. Furthermore, we uncover that spatial income segregation plays a key role in model transferability: mobility networks are most reliably reconstructed when target cities share similar segregation levels with the source. We design an index to quantify this segregation and accurately predict transferability. Finally, we generate mobility flow proxies for over 1,200 cities worldwide, highlighting neuroGravity's potential to mitigate critical data shortages in resource-limited, underdeveloped areas.