mlsanityData to 14 Oct 2026

Contrastive Pretraining for Visual Concept Explanations of Socioeconomic Outcomes

arXiv 2404.09768 ↗First posted 15 Apr 2024cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 16 of its 33 references that are in the map. First posted on arXiv 15 Apr 2024.

Builds on

16 of 33 references.

References · 16

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  3. 2019RandAugment: Practical automated data augmentation with a reduced search spacecited by 2,269
  4. 2016Understanding intermediate layers using linear classifier probescited by 1,353
  5. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,319
  6. 2017Random Erasing Data Augmentationcited by 1,239
  7. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 708
  8. 2020Concept Whitening for Interpretable Image Recognitioncited by 225
  9. 2022From Attribution Maps to Human-Understandable Explanations through Concept Relevance Propagationcited by 141
  10. 2017Interpreting Deep Visual Representations via Network Dissectioncited by 103
  11. 2022Rank-N-Contrast: Learning Continuous Representations for Regressioncited by 76
  12. 2020Using satellite imagery to understand and promote sustainable developmentcited by 71
4 more
  1. 2022"Help Me Help the AI": Understanding How Explainability Can Support Human-AI Interactioncited by 63
  2. 2022Overlooked factors in concept-based explanations: Dataset choice, concept learnability, and human capabilitycited by 46
  3. 2023FLAIR: a Country-Scale Land Cover Semantic Segmentation Dataset From Multi-Source Optical Imagerycited by 41
  4. 2018Local Explanation Methods for Deep Neural Networks Lack Sensitivity to Parameter Valuescited by 40

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Chronological Contrastive Learning: Few-Shot Progression Assessment in Irreversible Diseasescited by 0