{"id":"W2986406282","doi":"10.1080/22020586.2019.12072978","title":"Using machine learning to interpret 3D airborne electromagnetic inversions","year":2019,"lang":"en","type":"article","venue":"ASEG Extended Abstracts","topic":"Geophysical and Geoelectrical Methods","field":"Earth and Planetary Sciences","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western Forest Products; BC Hydro (Canada); Geoscience BC; University of British Columbia Hospital","funders":"","keywords":"Inversion (geology); A priori and a posteriori; Computer science; Algorithm; Artificial intelligence; Pixel; Synthetic data; Regularization (linguistics); Interpretation (philosophy); Machine learning; Geology","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":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0002392541,0.0001853068,0.0002349713,0.0001189687,0.0001259683,0.00005582442,0.0002240487,0.00007754591,0.003700004],"category_scores_gemma":[0.0002581618,0.0001532761,0.00008095889,0.0004818744,0.00002519228,0.0001599359,0.00002652361,0.0004406086,0.003211183],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000807603,"about_ca_system_score_gemma":0.00004630177,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002376449,"about_ca_topic_score_gemma":0.00024413,"domain_scores_codex":[0.9984441,0.0001286807,0.0002345913,0.0003809407,0.0002655046,0.0005461944],"domain_scores_gemma":[0.9990122,0.0003561281,0.00005459367,0.0001993999,0.00003658115,0.0003411298],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0001153896,0.00007155567,0.002321316,0.00002002966,0.00002532542,0.00004153519,0.00015115,0.01604175,0.007054241,0.00005213972,0.0001134771,0.9739921],"study_design_scores_gemma":[0.0002755424,0.001039261,0.9308954,0.00003105153,0.00002411268,0.00002395849,0.00003361404,0.05859554,0.003255058,0.00174985,0.003763537,0.0003130932],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9921698,0.0001897346,0.000134747,0.0002579445,0.0002422344,0.0002182831,0.00001453712,0.00007996229,0.006692789],"genre_scores_gemma":[0.9839315,0.00001013933,0.01438612,0.000686225,0.00008247289,7.258436e-7,0.0000415682,0.000006690787,0.0008545826],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.973679,"threshold_uncertainty_score":0.9975649,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01582657005552447,"score_gpt":0.2514695281012532,"score_spread":0.2356429580457287,"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."}}