{"id":"W3137738592","doi":"10.1111/mice.12655","title":"A knowledge‐enhanced deep reinforcement learning‐based shape optimizer for aerodynamic mitigation of wind‐sensitive structures","year":2021,"lang":"en","type":"article","venue":"Computer-Aided Civil and Infrastructure Engineering","topic":"Wind and Air Flow Studies","field":"Environmental Science","cited_by":78,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec à Chicoutimi","funders":"","keywords":"Reinforcement learning; Aerodynamics; Computer science; Artificial neural network; Artificial intelligence; Domain knowledge; Shape optimization; Process (computing); Domain (mathematical analysis); Deep learning; Machine learning; Engineering; Aerospace engineering; Mathematics; Structural engineering; Finite element method","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":[0.00006902919,0.000240639,0.0003034918,0.00005057749,0.0001307471,0.00003200015,0.00009292149,0.00008487469,0.0001440105],"category_scores_gemma":[0.00005318995,0.0002245212,0.00009391617,0.0001830181,0.00008016496,0.0001265377,0.0001668075,0.0001609952,0.000001836918],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006736792,"about_ca_system_score_gemma":0.00001725112,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000294688,"about_ca_topic_score_gemma":0.00001034367,"domain_scores_codex":[0.9989484,0.00002284801,0.0002704387,0.000343863,0.0001449445,0.0002695147],"domain_scores_gemma":[0.9995249,0.0001131096,0.00009403013,0.0001361257,0.00005147042,0.00008040144],"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":[0.00001271351,0.000008771624,0.000225715,0.00009675733,0.0000430775,0.000002105641,0.0006371738,0.9514874,0.03790998,0.00009956731,0.0001183716,0.009358411],"study_design_scores_gemma":[0.0007711669,0.0001213914,0.05556946,0.00009014057,0.00002932749,0.000006926886,0.00006994226,0.9228418,0.01957869,0.000128063,0.0005407307,0.0002523491],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5070121,0.000114808,0.4920579,0.00003154864,0.0002474199,0.0001648538,0.000003332145,0.00004947283,0.0003185789],"genre_scores_gemma":[0.9729631,0.00001772019,0.02669084,0.00006602212,0.0001446549,0.00001001246,0.00003945746,0.00002076346,0.00004745051],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.465951,"threshold_uncertainty_score":0.9155707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.003787149646000442,"score_gpt":0.19159702718897,"score_spread":0.1878098775429696,"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."}}