https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/Head https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://www.nanopub.org/nschema#hasAssertion https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/assertion https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://www.nanopub.org/nschema#hasProvenance https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/provenance https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://www.nanopub.org/nschema#hasPublicationInfo https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/pubinfo https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://www.w3.org/1999/02/22-rdf-syntax-ns#type http://www.nanopub.org/nschema#Nanopublication https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/assertion https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome http://schema.org/endDate 2026-04-17 https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome http://www.w3.org/1999/02/22-rdf-syntax-ns#type https://w3id.org/sciencelive/o/terms/FORRT-Replication-Outcome https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome http://www.w3.org/2000/01/rdf-schema#label Supervised pretraining matches episodic meta-learning on EuroSAT with 12× less compute https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/hasConclusionDescription Standard supervised pretraining — training ResNet-10 to classify everyday objects for 10 epochs (~15 minutes) — achieves 76.2% accuracy on 5-way 5-shot Sentinel-2 satellite land cover classification. This is comparable to episodic meta-learning trained for 40,000 episodes (~3 hours) which achieves 75.0%. Supervised pretraining requires no meta-learning framework or expertise, only standard deep learning classification training, making cross-domain few-shot satellite classification accessible to researchers without specialised AI knowledge. https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/hasConfidenceLevel https://w3id.org/sciencelive/o/terms/HighConfidence https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/hasEvidenceDescription Supervised 10 epochs: 5-shot 76.2% ± 1.7%, 20-shot 79.5% ± 1.1%, 50-shot 81.5% ± 1.2%. Episodic 40,000 steps: 5-shot 75.0% ± 0.8%, 20-shot 81.9% ± 0.6%, 50-shot 82.9% ± 0.6%. Training time: supervised ~15 minutes, episodic ~3 hours (12× faster). Same ResNet-10 backbone, 224×224 images, same data augmentation. https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/hasLimitationsDescription Only 10 supervised epochs (full convergence at 400 epochs may improve results further). Wider confidence intervals (±1.7%) due to 100 evaluation episodes. Only RGB bands from EuroSAT used. Supervised approach may perform differently on target domains more dissimilar to photographs than satellite imagery. https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/hasOutcomeRepository https://doi.org/10.5281/zenodo.19643138 https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/hasValidationStatus https://w3id.org/sciencelive/o/terms/Validated https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/o/terms/isOutcomeOf https://w3id.org/sciencelive/np/RAW3U5bhevQdUMc51N_QsJSuCrV8j8QauGh9cU6H0o-2Q/supervised-pretraining https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/provenance https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/assertion http://www.w3.org/ns/prov#wasAttributedTo https://orcid.org/0000-0002-1784-2920 https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/pubinfo https://orcid.org/0000-0002-1784-2920 http://xmlns.com/foaf/0.1/name Anne Fouilloux https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://purl.org/dc/terms/created 2026-04-18T16:18:39.812Z https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://purl.org/dc/terms/creator https://orcid.org/0000-0002-1784-2920 https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://purl.org/dc/terms/license https://creativecommons.org/licenses/by/4.0/ https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://purl.org/nanopub/x/introduces https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/supervised-pretraining-outcome https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://purl.org/nanopub/x/wasCreatedAt https://platform.sciencelive4all.org https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc http://www.w3.org/2000/01/rdf-schema#label NP created using Declaring a replication study outcome according to FORRT https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc https://w3id.org/np/o/ntemplate/wasCreatedFromTemplate https://w3id.org/np/RA2zljn0Nw9SadppOyxZoh-_Rxosslrq-vYG-p9SttnJE https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/sig http://purl.org/nanopub/x/hasAlgorithm RSA https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/sig http://purl.org/nanopub/x/hasPublicKey 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 https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/sig http://purl.org/nanopub/x/hasSignature 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 https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/sig http://purl.org/nanopub/x/hasSignatureTarget https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc https://w3id.org/sciencelive/np/RA9PP1TVbvRwEv9UNHYfGWvfCXtxCe3f6_xJnf-YbAgSc/sig http://purl.org/nanopub/x/signedBy https://orcid.org/0000-0002-1784-2920