{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/impact-assessment-of-missing-data-in-model","title":"Impact Assessment of Missing Data in Model Predictions for Earth Observation Applications","arxiv_id":"2403.14297","date":"2024-03-21","proceeding":null,"authors":["Francisco Mena","Diego Arenas","Marcela Charfuelan","Marlon Nuske","Andreas Dengel"],"abstract":"Earth observation (EO) applications involving complex and heterogeneous data sources are commonly approached with machine learning models. However, there is a common assumption that data sources will be persistently available. Different situations could affect the availability of EO sources, like noise, clouds, or satellite mission failures. In this work, we assess the impact of missing temporal and static EO sources in trained models across four datasets with classification and regression tasks. We compare the predictive quality of different methods and find that some are naturally more robust to missing data. The Ensemble strategy, in particular, achieves a prediction robustness up to 100%. We evidence that missing scenarios are significantly more challenging in regression than classification tasks. Finally, we find that the optical view is the most critical view when it is missing individually.","url_abs":"https://arxiv.org/abs/2403.14297v2","url_pdf":"https://arxiv.org/pdf/2403.14297v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"impact-assessment-of-missing-data-in-model","repo_url":"https://github.com/fmenat/missingviews-study-eo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification-on-time-series-with-missing","task_name":"Classification on Time Series with Missing Data"},{"task_slug":"crop-classification","task_name":"Crop Classification"},{"task_slug":"earth-observation","task_name":"Earth Observation"},{"task_slug":"multi-view-learning","task_name":"MULTI-VIEW LEARNING"},{"task_slug":"sensor-fusion","task_name":"Sensor Fusion"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/crop-classification-on-cropharvest-global","task":"Crop Classification","dataset":"CropHarvest - Global","model":"Feature Gated Fusion","rank_in_archive_order":1,"of":3,"metrics":{"Average Accuracy":"0.849"},"uses_additional_data":false},{"leaderboard":"/sota/crop-classification-on-cropharvest-global","task":"Crop Classification","dataset":"CropHarvest - Global","model":"Input Fusion","rank_in_archive_order":2,"of":3,"metrics":{"Average Accuracy":"0.847"},"uses_additional_data":false},{"leaderboard":"/sota/crop-classification-on-cropharvest-global","task":"Crop Classification","dataset":"CropHarvest - Global","model":"Ensemble strategy","rank_in_archive_order":3,"of":3,"metrics":{"Average Accuracy":"0.828"},"uses_additional_data":false},{"leaderboard":"/sota/crop-classification-on-cropharvest-multicrop","task":"Crop Classification","dataset":"CropHarvest multicrop - Global","model":"Input Fusion","rank_in_archive_order":1,"of":3,"metrics":{"Average Accuracy":"0.738"},"uses_additional_data":false},{"leaderboard":"/sota/crop-classification-on-cropharvest-multicrop","task":"Crop Classification","dataset":"CropHarvest multicrop - Global","model":"Feature Gated Fusion","rank_in_archive_order":2,"of":3,"metrics":{"Average Accuracy":"0.734"},"uses_additional_data":false},{"leaderboard":"/sota/crop-classification-on-cropharvest-multicrop","task":"Crop Classification","dataset":"CropHarvest multicrop - Global","model":"Ensemble strategy","rank_in_archive_order":3,"of":3,"metrics":{"Average Accuracy":"0.715"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}