# Far from the Crowd: Scalable Self-Supervised Learning via Geographic Isolation

arXiv: https://arxiv.org/abs/2608.19766 · HTML: https://mlsanity.com/paper/2608.19766/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 47 references that are in the map.
- First posted on arXiv 20 Aug 2026.

## Builds on (most cited first, 30 of 36)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Improved Baselines with Momentum Contrastive Learning](https://mlsanity.com/paper/2003.04297/) (2020)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [Beyond neural scaling laws: beating power law scaling via data pruning](https://mlsanity.com/paper/2206.14486/) (2022)
- [Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics](https://mlsanity.com/paper/2009.10795/) (2020)
- [Understanding Dimensional Collapse in Contrastive Self-supervised Learning](https://mlsanity.com/paper/2110.09348/) (2021)
- [On The Power of Curriculum Learning in Training Deep Networks](https://mlsanity.com/paper/1904.03626/) (2019)
- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [Prioritized Training on Points that are Learnable, Worth Learning, and Not Yet Learnt](https://mlsanity.com/paper/2206.07137/) (2022)
- [Self-paced Contrastive Learning with Hybrid Memory for Domain Adaptive Object Re-ID](https://mlsanity.com/paper/2006.02713/) (2020)
- [Geography-Aware Self-Supervised Learning](https://mlsanity.com/paper/2011.09980/) (2020)
- [Self-supervised Learning in Remote Sensing: A Review](https://mlsanity.com/paper/2206.13188/) (2022)
- [Curriculum Learning by Transfer Learning: Theory and Experiments with Deep Networks](https://mlsanity.com/paper/1802.03796/) (2018)
- [Region Similarity Representation Learning](https://mlsanity.com/paper/2103.12902/) (2021)
- [Time for a change: a tutorial for comparing multiple classifiers through Bayesian analysis](https://mlsanity.com/paper/1606.04316/) (2016)
- [Dynamic Curriculum Learning for Imbalanced Data Classification](https://mlsanity.com/paper/1901.06783/) (2019)
- [SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observation](https://mlsanity.com/paper/2211.07044/) (2022)
- [Towards a Unified Copernicus Foundation Model for Earth Vision](https://mlsanity.com/paper/2503.11849/) (2025)
- [SSL4EO-L: Datasets and Foundation Models for Landsat Imagery](https://mlsanity.com/paper/2306.09424/) (2023)
- [reBEN: Refined BigEarthNet Dataset for Remote Sensing Image Analysis](https://mlsanity.com/paper/2407.03653/) (2024)
- [Data-Efficient Contrastive Self-supervised Learning: Most Beneficial Examples for Supervised Learning Contribute the Least](https://mlsanity.com/paper/2302.09195/) (2023)
- [Objectives Matter: Understanding the Impact of Self-Supervised Objectives on Vision Transformer Representations](https://mlsanity.com/paper/2304.13089/) (2023)
- [Learning Sample Difficulty from Pre-trained Models for Reliable Prediction](https://mlsanity.com/paper/2304.10127/) (2023)
- [Improving Self-supervised Learning with Hardness-aware Dynamic Curriculum Learning: An Application to Digital Pathology](https://mlsanity.com/paper/2108.07183/) (2021)

## Abstract (arXiv)

Self-supervised pretraining on remote sensing imagery typically treats all samples as equally informative, despite large variability in geographic and visual structure. We propose a curriculum learning strategy for self-supervised Earth observation that ranks samples by geographic isolation, a label-free proxy derived entirely from geolocation metadata already present in geospatial datasets, requiring no image decoding, no model feedback, and no manual annotation. Unlike visual complexity proxies, it scales as O(D log D) with dataset size D and is well-defined for both contrastive and reconstructive objectives. We integrate the proposed measure into MoCoV2 and MAE pretraining and evaluate across three downstream tasks from CopernicusBench (BigEarthNet, DFC-2020, LCZ). Our curriculum reaches baseline final-epoch performance using as few as 20% of the training budget (MAE) and at most 40% (MoCo) of the training budget, and improves final downstream performance by up to +5 mAP on BigEarthNet, with gains of 1-5 points across benchmarks, matching visual-complexity curricula while reducing pre-computation cost by more than 140x (4 s vs. 568 s on SSL4EO). A CKA and effective-rank analysis further reveals that curriculum-trained encoders develop higher-dimensional, more uniformly utilized embedding spaces throughout training.