{"id":"W4365460520","doi":"10.48550/arxiv.2304.05509","title":"Control invariant set enhanced reinforcement learning for process control: improved sampling efficiency and guaranteed stability","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Control Systems Optimization","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Backup; Computer science; Stability (learning theory); Sampling (signal processing); Process (computing); Invariant (physics); State (computer science); Artificial intelligence; Control theory (sociology); Control (management); Machine learning; Mathematical optimization; Algorithm; Mathematics; Detector","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005698927,0.0004710535,0.0007261263,0.0002089634,0.0002210055,0.00008322984,0.0003495548,0.0003424361,0.000009493098],"category_scores_gemma":[0.0003815294,0.0005607894,0.0001667317,0.0002834363,0.00007480472,0.0002127384,0.0001278834,0.0005594584,0.000005510642],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003496545,"about_ca_system_score_gemma":0.00009551134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006334066,"about_ca_topic_score_gemma":0.00004714501,"domain_scores_codex":[0.9978352,0.00008853655,0.0005249285,0.0009024832,0.00009353447,0.0005552782],"domain_scores_gemma":[0.9982869,0.0004628559,0.0003314096,0.0004797451,0.0003000346,0.0001390571],"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":[0.0003158916,0.0000120319,0.000251139,0.0008112693,0.0002347205,0.000004824693,0.0003602802,0.9934479,0.003771492,0.0007280766,0.000001514763,0.00006083162],"study_design_scores_gemma":[0.004649812,0.00009840883,0.00007748734,0.0002000639,0.0002101506,8.188259e-7,0.0003435469,0.9920288,0.0004008839,0.001438366,0.00001826166,0.000533344],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1514665,0.0000633724,0.8447823,0.00001207711,0.000370913,0.002371314,0.0000693089,0.0007799504,0.00008433152],"genre_scores_gemma":[0.9992248,0.00005899893,0.0002760734,0.00001917951,0.00008590668,0.00006128124,0.00008251442,0.00009436542,0.00009681293],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8477584,"threshold_uncertainty_score":0.9996843,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05053102258716711,"score_gpt":0.2040948357973252,"score_spread":0.1535638132101581,"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."}}