{"id":"W3092762910","doi":"10.5194/hess-25-2045-2021","title":"Rainfall–runoff prediction at multiple timescales with a single Long Short-Term Memory network","year":2021,"lang":"en","type":"article","venue":"Hydrology and earth system sciences","topic":"Hydrological Forecasting Using AI","field":"Environmental Science","cited_by":253,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Global Water Futures; Horizon 2020 Framework Programme; Janssen Pharmaceuticals; Canada First Research Excellence Fund; Österreichische Forschungsförderungsgesellschaft; European Commission; Bundesministerium für Bildung, Wissenschaft und Forschung; Google; Compute Canada; Nvidia","keywords":"Benchmark (surveying); Process (computing); Temporal resolution; Flood myth; Deep learning; Reservoir computing; Long short term memory; Memory model","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007465091,0.0001728633,0.0002454675,0.00002659238,0.0009428106,0.00006251218,0.0001721526,0.0001426026,0.0003093722],"category_scores_gemma":[0.00004192883,0.0001243357,0.00003882589,0.0003750316,0.001382367,0.0001961072,0.0002576542,0.0001245653,0.0001343551],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000455493,"about_ca_system_score_gemma":0.00001931138,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004988206,"about_ca_topic_score_gemma":0.001311398,"domain_scores_codex":[0.9980245,0.0002268668,0.0002426681,0.0006607619,0.0003162925,0.000528917],"domain_scores_gemma":[0.9993908,0.0001714167,0.00008011379,0.0001951888,0.0000111453,0.0001513511],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00004668952,0.00004421957,0.8445187,0.00002182503,0.00001585326,0.000156925,0.0002302588,0.1484379,0.00435668,0.00002700704,0.0001992014,0.001944772],"study_design_scores_gemma":[0.0008974529,0.001668416,0.5663097,0.0002686165,0.00008922396,0.004054275,0.0001847616,0.4203535,0.003438668,0.00009701389,0.001990368,0.0006479545],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.984445,0.0002497855,0.0003153053,0.0001880757,0.0002508305,0.0001526628,0.000004220013,0.0001343359,0.01425983],"genre_scores_gemma":[0.9979444,0.00000616904,0.001089538,0.0001692091,0.0001033788,0.00001226906,0.00000781418,0.000007721297,0.0006594561],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2782089,"threshold_uncertainty_score":0.7251432,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0189434626163235,"score_gpt":0.202268328648595,"score_spread":0.1833248660322715,"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."}}