{"id":"W3212195610","doi":"10.1088/2632-2153/ac3845","title":"Gradient domain machine learning with composite kernels: improving the accuracy of PES and force fields for large molecules","year":2021,"lang":"en","type":"article","venue":"Machine Learning Science and Technology","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Ab initio; Generalization; Computer science; Metric (unit); Kernel (algebra); Energy minimization; Force field (fiction); Algorithm; Artificial intelligence; Machine learning; Mathematics; Computational chemistry; Physics; Chemistry; Mathematical analysis","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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.003129949,0.0002614001,0.0003936644,0.0002824384,0.001651669,0.0003164475,0.0007365129,0.000116357,0.00004671384],"category_scores_gemma":[0.003360299,0.0001745797,0.00003388058,0.00105202,0.001563982,0.0003182407,0.0008540617,0.0005704356,0.000002960763],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003157922,"about_ca_system_score_gemma":0.0001704383,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002117285,"about_ca_topic_score_gemma":0.0001445796,"domain_scores_codex":[0.9973574,0.0002007171,0.0003666978,0.0008763552,0.0004979069,0.0007009297],"domain_scores_gemma":[0.9981663,0.0005169998,0.0004123287,0.0004315257,0.0003792952,0.00009350148],"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":[0.00005133791,0.00003889216,0.03736322,0.0001168178,0.000007661461,0.00001582627,0.0008920097,0.001444272,0.9404539,0.01487764,0.000004088191,0.004734308],"study_design_scores_gemma":[0.002509787,0.002046819,0.008325622,0.0002574196,0.00009496905,0.0009763864,0.002881821,0.3667091,0.600131,0.008202031,0.006939884,0.0009251114],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9691341,0.001181552,0.02553773,0.003365336,0.0001233079,0.000306934,0.00001454714,0.0001916896,0.0001447926],"genre_scores_gemma":[0.978482,0.00006926528,0.02096318,0.0001633779,0.00002575634,0.00004117511,0.000007119332,0.00002209737,0.0002260227],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3652648,"threshold_uncertainty_score":0.999648,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005665228847731934,"score_gpt":0.2400752758569117,"score_spread":0.2344100470091797,"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."}}