{"id":"W7005478051","doi":"","title":"Regularized reinforcement learning with performance guarantees","year":2014,"lang":"en","type":"dissertation","venue":"eScholarship@McGill (McGill)","topic":"Chemistry and Stereochemistry Studies","field":"Chemistry","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"McGill University","keywords":"Reinforcement learning; Curse of dimensionality; Leverage (statistics); Regularization (linguistics); Bayesian probability; Domain knowledge; Correctness","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","research_integrity","insufficient_payload"],"consensus_categories":["metaepi_narrow"],"category_scores_codex":[0.00045392,0.001325099,0.001241646,0.000107317,0.001871175,0.000125576,0.0008774981,0.001049741,0.001329439],"category_scores_gemma":[0.0004563705,0.001269601,0.0004043774,0.0002920534,0.0001349719,0.0003744631,0.0001862101,0.002701886,0.0001555047],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005545005,"about_ca_system_score_gemma":0.00006860594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005361825,"about_ca_topic_score_gemma":0.00007852717,"domain_scores_codex":[0.9951951,0.00006206788,0.001080233,0.001456706,0.001164615,0.001041303],"domain_scores_gemma":[0.9968043,0.0001911099,0.001040733,0.001168016,0.0004964894,0.0002993847],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002708483,0.0001975873,0.000301301,0.009295524,0.001387313,0.0001210876,0.00004298848,0.0006916831,0.9357348,0.001443556,0.00002450037,0.04805118],"study_design_scores_gemma":[0.002057674,0.0001349753,0.00008691236,0.002317618,0.0004166138,0.0000677336,0.0005992265,0.0001540105,0.8979109,0.0002423962,0.09431512,0.001696814],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6218174,0.0001411814,1.486185e-7,0.000005467331,0.0001473428,0.000134499,0.00007463559,0.0003119627,0.3773673],"genre_scores_gemma":[0.8139458,0.0003554382,0.0001625948,0.00004396746,0.0001296254,0.0001943578,0.00250337,0.0001944347,0.1824704],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.194897,"threshold_uncertainty_score":0.9999501,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009356948074453714,"score_gpt":0.2102365491806401,"score_spread":0.2008796011061864,"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."}}