{"id":"W4225553680","doi":"10.1175/mwr-d-21-0154.1","title":"A hybrid analog-ensemble, convolutional-neural-network method for post-processing precipitation forecasts","year":2022,"lang":"en","type":"article","venue":"Monthly Weather Review","topic":"Meteorological Phenomena and Simulations","field":"Earth and Planetary Sciences","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"BC Hydro (Canada); University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; BC Hydro; Mitacs; University Corporation for Atmospheric Research; University of British Columbia; Nvidia; National Center for Atmospheric Research; National Science Foundation","keywords":"Precipitation; Convolutional neural network; Quantitative precipitation forecast; Computer science; Artificial neural network; Meteorology; Environmental science; Artificial intelligence; Physics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001391316,0.0001652825,0.0003585112,0.00004550117,0.0006516612,0.00003532635,0.0002242908,0.00001966424,0.003811859],"category_scores_gemma":[0.0001634106,0.0001309968,0.0001819853,0.0002858939,0.00002711349,0.000158063,0.00002340906,0.0001462689,0.00003625824],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001597914,"about_ca_system_score_gemma":0.00005159807,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008180096,"about_ca_topic_score_gemma":0.00009163241,"domain_scores_codex":[0.998131,0.0004470476,0.0004022029,0.0003546458,0.0002838512,0.0003813009],"domain_scores_gemma":[0.9988896,0.0005100149,0.0001958396,0.0001848835,0.0001073897,0.0001122552],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009942184,0.00004278473,0.004612151,0.0005000919,0.00004618338,0.00000824438,0.0001016502,0.4595842,0.000009280903,0.0006143405,0.01014603,0.5242356],"study_design_scores_gemma":[0.0004089092,0.0005628468,0.01999576,0.0001660112,0.0001656404,0.00002068793,0.00003393591,0.6699154,0.0000013654,0.02541874,0.2829618,0.0003489318],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"review","genre_gemma":"empirical","genre_scores_codex":[0.05393384,0.8288239,0.07320081,0.009627841,0.001197311,0.006322807,0.002353003,0.0003259806,0.02421454],"genre_scores_gemma":[0.9231757,0.001656306,0.05351887,0.0164746,0.0003701788,0.0003098345,0.003067436,0.0000235821,0.001403535],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8692418,"threshold_uncertainty_score":0.9970988,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04079427134220216,"score_gpt":0.2826245887955047,"score_spread":0.2418303174533025,"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."}}