Contrastive Pretraining for Visual Concept Explanations of Socioeconomic Outcomes

MCML Authors

Ivica Obadic

→ Group Xiaoxiang Zhu
 Data Science in Earth Observation

Lars Pennig

→ Group Niki Kilbertus
 Ethics in Systems Design and Machine Learning

Xiaoxiang Zhu

Prof. Dr.

Principal Investigator

Data Science in Earth Observation

Abstract

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 urban concepts 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.

inproceedings

---

Workshop @CVPR 2024

Workshop at the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, WA, USA, Jun 17-21, 2024.

---

Authors

I. Obadic • A. Levering • L. Pennig • D. Oliveira • D. Marcos • X. Zhu

---

Links

DOI

---

Research Areas

 A3 | Computational Models

 C3 | Physics and Geo Sciences