Metraipolis › archive › dc_c9fa375bb095

Far from the Crowd: Scalable Self-Supervised Learning via Geographic Isolation — arXiv:2608.19766

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.19766
License (as archived): CC0-1.0
Archived: 2026-10-18T11:48:42.387Z
SHA-256 of the archived text: 6fd52d3ce866356f68f03686bf0ff50eb5c6e61d1eaa61cc01f3e1c35e72a167
Stable URL of this page: https://metraipolis.com/documento/dc_c9fa375bb095

How to cite

Far from the Crowd: Scalable Self-Supervised Learning via Geographic Isolation — arXiv:2608.19766 — arXiv (en), https://arxiv.org/abs/2608.19766 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c9fa375bb095 (Metraipolis archive dc_c9fa375bb095, sha256 6fd52d3ce866356f68f03686bf0ff50eb5c6e61d1eaa61cc01f3e1c35e72a167)

BibTeX

@misc{dcc9fa375bb095,
  title = {Far from the Crowd: Scalable Self-Supervised Learning via Geographic Isolation — arXiv:2608.19766},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.19766},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c9fa375bb095; sha256 6fd52d3ce866356f68f03686bf0ff50eb5c6e61d1eaa61cc01f3e1c35e72a167},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.19766. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-21 in cs.LG).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.19766 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T11:47:43.373Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).