Contrastive Pretraining for Visual Concept Explanations of Socioeconomic Outcomes
This work addresses the need for interpretable models in policy-making by enabling clearer visual concept explanations for socioeconomic predictions, though it is incremental as it builds on existing contrastive learning and explainability methods.
The paper tackled the problem of improving interpretability in deep learning models that predict socioeconomic indicators from satellite imagery by using task-specific contrastive pretraining, resulting in a latent space that orders embeddings according to socioeconomic outcomes and enhances concept-based explanations.
Predicting socioeconomic indicators from satellite imagery with deep learning has become an increasingly popular research direction. Post-hoc concept-based explanations can be an important step towards broader adoption of these models in policy-making as they enable the interpretation of socioeconomic outcomes based on visual concepts that are intuitive to humans. In this paper, we study the interplay between representation learning using an additional task-specific contrastive loss and post-hoc concept explainability for socioeconomic studies. Our results on two different geographical locations and tasks indicate that the task-specific pretraining imposes a continuous ordering of the latent space embeddings according to the socioeconomic outcomes. This improves the model's interpretability as it enables the latent space of the model to associate concepts encoding typical urban and natural area patterns with continuous intervals of socioeconomic outcomes. Further, we illustrate how analyzing the model's conceptual sensitivity for the intervals of socioeconomic outcomes can shed light on new insights for urban studies.