{"id":"W3170198215","doi":"10.5194/egusphere-egu21-12931","title":"Calibration of a reinforcement learning method with the ClimEx large ensemble and a weather generator for water management ","year":2021,"lang":"en","type":"article","venue":"","topic":"Water resources management and optimization","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Rio Tinto (Canada); École de Technologie Supérieure","funders":"","keywords":"Streamflow; Reinforcement learning; Computer science; Hydropower; Calibration; Time horizon; Hydrological modelling; Environmental science; Artificial intelligence; Mathematical optimization; Drainage basin; Mathematics; Engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.003019781,0.0007343006,0.0005889356,0.0005367737,0.0003599689,0.0006812396,0.001007234,0.001147766,0.001486325],"category_scores_gemma":[0.008230651,0.0003764055,0.000448818,0.0003781985,0.0005763035,0.0007086188,0.0008781993,0.001624321,0.0002371867],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001023621,"about_ca_system_score_gemma":0.001017381,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01197319,"about_ca_topic_score_gemma":0.005513265,"domain_scores_codex":[0.9993417,0.0003180432,0.00003347151,0.0001479284,0.000096558,0.00006227366],"domain_scores_gemma":[0.9970303,0.001807538,0.0002248797,0.0003481568,0.0004843801,0.0001048052],"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":[0.00008338878,0.00008149329,0.001680484,0.00002488967,0.00003906188,0.00002949828,0.00002324854,0.9785484,0.0005719439,0.0007589576,0.0005611861,0.01759743],"study_design_scores_gemma":[0.000009303882,0.00001558389,0.0003735621,0.000002784082,0.000002229161,0.000003414676,0.000003363549,0.9988137,0.0004194588,0.0002231509,0.0001296353,0.000003811389],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5428675,0.0005176208,0.4458093,0.0006268303,0.0001599873,0.0002788457,0.0007951925,0.002929654,0.006014958],"genre_scores_gemma":[0.9447553,0.00006259284,0.05344224,0.00007632284,0.00001574864,0.000164229,0.0006690707,0.00007071133,0.0007437675],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01197319,"threshold_uncertainty_score":0.02380699,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00997145771974011,"score_gpt":0.2111902625617636,"score_spread":0.2012188048420235,"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."}}