{"id":"W4377100951","doi":"10.1016/j.jhydrol.2023.129682","title":"Applying transfer learning techniques to enhance the accuracy of streamflow prediction produced by long Short-term memory networks with data integration","year":2023,"lang":"en","type":"article","venue":"Journal of Hydrology","topic":"Hydrology and Watershed Management Studies","field":"Environmental Science","cited_by":51,"is_retracted":false,"has_abstract":false,"ca_institutions":"Environment and Climate Change Canada; Institut National de la Recherche Scientifique","funders":"Natural Sciences and Engineering Research Council of Canada; Québec Ministère du Développement Durable, de l’Environnement et de la Lutte Contre les Changements Climatiques; Institut national de la recherche scientifique","keywords":"Streamflow; Computer science; Environmental science; Transfer of learning; Snow; Term (time); Transfer (computing); Artificial intelligence; Meteorology","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":[0.001044823,0.0001051027,0.0001923067,0.00006815045,0.0001555242,0.000009594823,0.0004006607,0.00005985412,0.00004939085],"category_scores_gemma":[0.00007173689,0.00006356421,0.00002730091,0.0002402958,0.0001550827,0.0003157015,0.0001892698,0.0003267041,0.000007285402],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002709598,"about_ca_system_score_gemma":0.000005097526,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001944311,"about_ca_topic_score_gemma":0.0000745615,"domain_scores_codex":[0.9989555,0.0001379206,0.0003152697,0.0002087011,0.0001846761,0.0001978975],"domain_scores_gemma":[0.9994881,0.0001113631,0.0001172335,0.0002359481,0.00001540257,0.0000319238],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001088323,0.0002050671,0.2232826,0.00004037672,0.0004697085,0.00008572634,0.003374841,0.3090797,0.2584596,0.000006517374,0.01998415,0.1839234],"study_design_scores_gemma":[0.001782144,0.01095226,0.1565717,0.0005566331,0.001348578,0.0004117312,0.002278263,0.1327748,0.6669008,0.0003334814,0.02486406,0.00122562],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9612424,0.00004561846,0.03605666,0.001900008,0.00009699804,0.0004145088,0.000002768106,0.00003527147,0.000205761],"genre_scores_gemma":[0.9991563,0.0002479876,0.0001805917,0.0001703371,0.00007433462,0.00004070298,0.0000166131,0.00000941995,0.00010371],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4084412,"threshold_uncertainty_score":0.2592073,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01505041570199334,"score_gpt":0.2696800649271941,"score_spread":0.2546296492252008,"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."}}