{"id":"W4390615242","doi":"10.26434/chemrxiv-2024-bpd53","title":"Leveraging our Teacher’s Experience to Improve Machine Learning: Application to pKa Prediction","year":2024,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Memorization; Leverage (statistics); Electronegativity; Computer science; Machine learning; Artificial intelligence; Quality (philosophy); Chemistry; Mathematics education; Mathematics; Epistemology; Organic chemistry","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002786844,0.0008270844,0.000523914,0.0005579386,0.000505096,0.001265583,0.00127853,0.001233857,0.002842173],"category_scores_gemma":[0.01698665,0.0002661546,0.0003330685,0.0006789697,0.0009347083,0.001873738,0.001901482,0.001838406,0.0009328054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005908419,"about_ca_system_score_gemma":0.000874864,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00212148,"about_ca_topic_score_gemma":0.002265963,"domain_scores_codex":[0.9989329,0.0005964735,0.000038145,0.0002139901,0.0001618392,0.00005666232],"domain_scores_gemma":[0.9921632,0.005191654,0.0004003811,0.0008850549,0.0009226539,0.0004370554],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006585356,0.002026068,0.0316663,0.0008479262,0.0001234282,0.0006588999,0.002810911,0.07377373,0.02168714,0.01353655,0.01478385,0.8374266],"study_design_scores_gemma":[0.0004465898,0.002768784,0.01134865,0.0003075429,0.0002235174,0.001033096,0.001578698,0.762272,0.0817444,0.06703676,0.07103484,0.0002051258],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4641147,0.006563622,0.470144,0.01637412,0.0007149513,0.0002570416,0.0003849105,0.004504738,0.03694191],"genre_scores_gemma":[0.8193247,0.001908221,0.1723415,0.0007203434,0.0001949294,0.00008164105,0.0001701516,0.0001947114,0.005063903],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002842173,"threshold_uncertainty_score":0.01473844,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03015043068578297,"score_gpt":0.328151508623609,"score_spread":0.298001077937826,"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."}}