$GRU^{spa}$: Gated Recurrent Unit with Spatial Attention for Spatio-Temporal Disaggregation

arXiv:2306.07292v3 Announce Type: replace-cross
Abstract: Open data is frequently released spatially aggregated, usually to comply with privacy policies. But coarse, heterogeneous aggregations complicate learning and integration for downstream AI/ML systems. In this work, we consider models to disaggregate spatio-temporal data from a low-resolution, irregular partition (e.g., census tract) to a high-resolution, irregular partition (e.g., city block). We propose a model, Gated Recurrent Unit with Spatial Attention ($GRU^{spa}$), where spatial attention layers are integrated into the original Gated Recurrent Unit (GRU) …

arxiv attention census cs.ai cs.cr cs.lg data gru integration low policies privacy privacy policies resolution spa systems temporal work