{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000368558,0.0001133766,0.0001579789,0.0001518229,0.0001431346,0.00008103881,0.00004865529,0.00004904748,0.00003214781],"category_scores_gemma":[0.00009372692,0.00009953333,0.00003546809,0.0001174339,0.0000110579,0.0001364331,0.000006596299,0.0001534791,0.00000198376],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008494727,"about_ca_system_score_gemma":0.00003099558,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002541615,"about_ca_topic_score_gemma":5.292698e-7,"domain_scores_codex":[0.9994035,0.00003490167,0.0002351991,0.0000873418,0.00007984081,0.0001592015],"domain_scores_gemma":[0.9994095,0.0002691667,0.00004446162,0.0000617729,0.0001523932,0.0000626804],"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":[0.00005321529,0.000003168217,0.0000723899,0.00002529946,0.0000762538,5.25332e-7,0.0002876603,0.9969403,0.000855061,0.0002099228,0.0002928266,0.001183349],"study_design_scores_gemma":[0.001591288,0.00007506584,0.0001305739,0.0001130292,0.00002365543,0.000007365318,0.0001143546,0.9951925,0.0002556237,0.00007798721,0.002309931,0.0001085713],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005704597,0.0003454017,0.9877095,0.0000598159,0.0001523616,0.0001757045,5.691365e-7,0.000096406,0.005755689],"genre_scores_gemma":[0.3470227,0.0006578243,0.6495764,0.00002182023,0.0002580379,0.00002941697,0.00001043725,0.00004571213,0.002377663],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.3413181,"threshold_uncertainty_score":0.4058851,"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."}}