{"id":"W3206564585","doi":"10.1029/2021gl094407","title":"Forecasting the Indian Ocean Dipole With Deep Learning Techniques","year":2021,"lang":"en","type":"article","venue":"Geophysical Research Letters","topic":"Oceanographic and Atmospheric Processes","field":"Earth and Planetary Sciences","cited_by":50,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Northern British Columbia","funders":"National Natural Science Foundation of China","keywords":"Convolutional neural network; Predictability; Deep learning; Computer science; Artificial intelligence; Anomaly (physics); Indian Ocean Dipole; Machine learning; Climatology; Sea surface temperature; Geology; Mathematics; Statistics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004438279,0.0001184421,0.0001272304,0.00002634493,0.0007642476,0.0002237148,0.0003246524,0.00003559063,0.0002849479],"category_scores_gemma":[0.0002542977,0.00006927471,0.00005273912,0.001151268,0.0004841499,0.0002008505,0.00004769195,0.0007964781,0.00009131332],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000004848954,"about_ca_system_score_gemma":0.00008399519,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007702609,"about_ca_topic_score_gemma":0.0002937784,"domain_scores_codex":[0.9979582,0.0002492428,0.0001094436,0.0003083744,0.0007169325,0.0006577802],"domain_scores_gemma":[0.9988214,0.0006785565,0.0000348265,0.0001979637,0.0001214591,0.0001457667],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001162957,0.00007430354,0.5780462,0.0001276916,0.0001101695,0.001187638,0.00193071,0.0008475736,0.001140023,0.000180109,0.003049252,0.41319],"study_design_scores_gemma":[0.001239404,0.002365027,0.7562472,0.000666759,0.00008232567,0.0004069702,0.02208626,0.03525854,0.02347802,0.009160257,0.1470517,0.001957492],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9909662,0.0002787306,0.0002591466,0.004894856,0.00002846485,0.0001215069,0.00000306461,0.00007527374,0.003372753],"genre_scores_gemma":[0.9970954,0.00004342777,0.001314119,0.0009214963,0.000280275,0.000002028071,0.00002836882,0.000006847902,0.000308019],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4112325,"threshold_uncertainty_score":0.5878052,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02326678836642944,"score_gpt":0.2515311713689399,"score_spread":0.2282643830025104,"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."}}