{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005877281,0.001882731,0.001724778,0.0006799541,0.0006094548,0.00139441,0.001903566,0.002298767,0.006419319],"category_scores_gemma":[0.01791731,0.000547912,0.001038755,0.0005499272,0.001437022,0.001805097,0.00214458,0.003237108,0.001118128],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002439964,"about_ca_system_score_gemma":0.004214693,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006996732,"about_ca_topic_score_gemma":0.005634768,"domain_scores_codex":[0.9967368,0.001384945,0.0001337386,0.0005566772,0.0006572918,0.000530443],"domain_scores_gemma":[0.9881003,0.0082756,0.0005316726,0.001361458,0.001171703,0.0005592728],"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":[0.0005388676,0.0001949949,0.0005807665,0.0001007948,0.00005493761,0.00002527567,0.00002285345,0.9576221,0.000599578,0.008250177,0.004171393,0.02783817],"study_design_scores_gemma":[0.0000559312,0.00006282795,0.0001204143,0.000009385801,0.000008229539,0.000008523551,0.000003854769,0.9943346,0.0003623542,0.004867718,0.0001618407,0.00000442228],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1678453,0.00222744,0.793619,0.003582334,0.0005209193,0.0003245293,0.0009406916,0.00471291,0.02622682],"genre_scores_gemma":[0.8792837,0.0002769105,0.1097948,0.0004771838,0.0001805665,0.0002399658,0.0007103928,0.0003694913,0.008666977],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006996732,"threshold_uncertainty_score":0.03108245,"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."}}