Context-Conditioned Generative Models Enable Subnational Refinement of Sparse Humanitarian Surveys
This work provides a method for humanitarian organizations to obtain more granular and accurate sub-national estimates from sparse survey data, which is crucial for effective decision-making and resource allocation.
This paper addresses the challenge of sparse survey data in humanitarian contexts by using context-conditioned generative models (normalizing flows) to refine sub-national survey distributions. They demonstrate that these models can improve sub-national estimates across eight household survey datasets from six low- and middle-income countries, with performance improving as conditioning information becomes richer.
Data scarcity limits inference in many scientific and policy domains. Survey data are essential for decision-making, but sparse samples often fail to capture fine spatial granularities. We evaluate normalizing flows, a generative model that learns complex data distributions and can be conditioned on exogenous contextual features, in controlled data scarcity scenarios. Across eight household survey datasets spanning six low-income or middle-income countries in the humanitarian domain, we show that context-conditioned generative models can refine sub-national survey distributions under severe data scarcity, and that performance increases systematically with the richness of the conditioning information. These findings support a general principle for survey data augmentation: generative models can improve sub-national estimates when the sparse sample retains sufficient support and contextual covariates encode relevant local heterogeneity. By learning full conditional distributions rather than point estimates, the approach provides fine-grained evidence for humanitarian decision-making and resource allocation.