{"id":"W7101391598","doi":"10.2118/231172-pa","title":"Imitation-Guided Reinforcement Learning with Adaptive Sampling and Expert Knowledge for Real-Time Well Control Optimization","year":2025,"lang":"en","type":"article","venue":"SPE Journal","topic":"Reservoir Engineering and Simulation Methods","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Reinforcement learning; Robustness (evolution); Benchmark (surveying); Convergence (economics); Artificial neural network; Process (computing); Adaptive sampling; Sampling (signal processing); Stability (learning theory)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001172017,0.000762525,0.0009476705,0.0003329539,0.0002514703,0.0006494246,0.0008997387,0.0007573161,0.00118904],"category_scores_gemma":[0.002680657,0.000415034,0.0004757765,0.0002192661,0.0007987185,0.0005997919,0.0008496451,0.001162653,0.0001751283],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006449912,"about_ca_system_score_gemma":0.0009442344,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006049431,"about_ca_topic_score_gemma":0.003770655,"domain_scores_codex":[0.9996642,0.0001032609,0.00002092883,0.00006895856,0.00008224007,0.00006044493],"domain_scores_gemma":[0.9989322,0.0006102861,0.0001397115,0.00007901976,0.0001722998,0.00006655959],"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":[0.00003595439,0.00002724812,0.0003581765,0.00002561573,0.00001851151,0.00003846979,0.00002148213,0.9818953,0.0009312137,0.001651178,0.0002394274,0.0147574],"study_design_scores_gemma":[0.000003578221,0.00001012554,0.00002203849,0.000001394726,0.00000186406,0.000002067744,0.00000105159,0.9994109,0.0001159457,0.0003765506,0.00005331532,0.000001249236],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05915984,0.0005076352,0.9362572,0.0002832355,0.00005830097,0.00004627786,0.00002264612,0.0007067511,0.002958061],"genre_scores_gemma":[0.9668359,0.00009221424,0.03181115,0.00008615188,0.00002102701,0.00006476478,0.00002775181,0.00003274659,0.001028316],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006049431,"threshold_uncertainty_score":0.0120284,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0227897904460014,"score_gpt":0.3005145443930562,"score_spread":0.2777247539470548,"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."}}