{"id":"W4321489725","doi":"10.5194/egusphere-egu23-4350","title":"Physics-Constrained Deep Learning for Downscaling","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Cryospheric studies and observations","field":"Earth and Planetary Sciences","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Downscaling; Upsampling; Computer science; Weather Research and Forecasting Model; Field (mathematics); Deep learning; Artificial intelligence; Variety (cybernetics); Adaptation (eye); Machine learning; Artificial neural network; Climate model; Data science; Climate change; Image (mathematics); Meteorology; Geography; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007291979,0.0009207022,0.0006509012,0.0005119899,0.0003305064,0.0007609149,0.001247679,0.0008961752,0.003563156],"category_scores_gemma":[0.003896599,0.0004752233,0.0005826678,0.0007316501,0.0006637559,0.001357749,0.001583821,0.002144073,0.000688199],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000894692,"about_ca_system_score_gemma":0.001168019,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007377409,"about_ca_topic_score_gemma":0.009314957,"domain_scores_codex":[0.9996849,0.00005772757,0.00002431248,0.00009022437,0.0001109745,0.00003183456],"domain_scores_gemma":[0.999378,0.0002593935,0.00007620025,0.0001438529,0.0001123989,0.0000301827],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000638596,0.00005944948,0.0008924709,0.0001590235,0.00007517995,0.0000708454,0.00004917633,0.8292485,0.006800468,0.02544215,0.006318762,0.1308201],"study_design_scores_gemma":[0.00000312651,0.000005325784,0.0001118639,0.000007080836,0.000003119651,0.00000760078,0.000002575855,0.9876368,0.001214257,0.01004947,0.0009551556,0.000003680706],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01483606,0.0005322756,0.9807075,0.0003555926,0.00006734796,0.0000348395,0.0003892398,0.001358294,0.001718917],"genre_scores_gemma":[0.4970326,0.001217377,0.4906483,0.0006710956,0.0001870459,0.000317892,0.003249903,0.0009259009,0.005749888],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007377409,"threshold_uncertainty_score":0.01466894,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.061898044868548,"score_gpt":0.2598801778268403,"score_spread":0.1979821329582923,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}