{"id":"W3131639001","doi":"10.1093/gji/ggab074","title":"Robust deep learning seismic inversion with<i>a priori</i>initial model constraint","year":2021,"lang":"en","type":"article","venue":"Geophysical Journal International","topic":"Seismic Imaging and Inversion Techniques","field":"Earth and Planetary Sciences","cited_by":88,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Key Research and Development Program of China; National Natural Science Foundation of China","keywords":"Inversion (geology); A priori and a posteriori; Seismic inversion; Synthetic data; Artificial neural network; Computer science; Seismic survey; Seismic to simulation; Reservoir modeling; Data set; Deep learning; Training set; Data mining; Algorithm; Geology; Artificial intelligence; Seismology; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000148832,0.0001325544,0.0001410335,0.00008356655,0.0002532869,0.0001976033,0.0002354618,0.00005183954,0.001472218],"category_scores_gemma":[0.00005560286,0.0001073591,0.0001038062,0.0001137335,0.0001251576,0.0004047145,0.00002788668,0.0006287934,0.0001685808],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002285086,"about_ca_system_score_gemma":0.0001702423,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001625196,"about_ca_topic_score_gemma":0.00001009616,"domain_scores_codex":[0.9987246,0.00007447777,0.0002067643,0.0002185707,0.0005360189,0.0002395761],"domain_scores_gemma":[0.9992928,0.00007872402,0.0001186763,0.00007884343,0.0002684894,0.0001624632],"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":[0.0002592879,0.0001390223,0.01666424,0.00001218047,0.0002026128,0.001029039,0.0004927997,0.5756026,0.0007969554,0.000798227,0.01082449,0.3931786],"study_design_scores_gemma":[0.0004603175,0.00009788921,0.001833586,0.00005026764,0.0000178194,0.0007508692,0.0003273432,0.9802908,0.00146815,0.003420979,0.01110951,0.0001724477],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7283198,0.0001423746,0.2120139,0.007101279,0.001412231,0.00008770727,0.00004166584,0.0001909453,0.05069013],"genre_scores_gemma":[0.984649,0.00008388425,0.0104093,0.00319584,0.0005110549,3.39524e-7,0.00009785817,0.000005518186,0.001047205],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4046883,"threshold_uncertainty_score":0.9994406,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01980427238500595,"score_gpt":0.2264958943487858,"score_spread":0.2066916219637799,"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."}}