{"id":"W2792351009","doi":"10.1063/1.5003074","title":"Neural networks vs Gaussian process regression for representing potential energy surfaces: A comparative study of fit quality and vibrational spectrum accuracy","year":2018,"lang":"en","type":"article","venue":"The Journal of Chemical Physics","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":243,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Government of Canada","keywords":"Kriging; Energy spectrum; Artificial neural network; Gaussian process; Quality (philosophy); Regression; Gaussian; Energy (signal processing); Process (computing); Statistics; Regression analysis; Spectrum (functional analysis); Artificial intelligence; Pattern recognition (psychology); Statistical physics; Machine learning; Computer science; Biological system; Mathematics; Physics; Biology; Quantum mechanics","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":[],"consensus_categories":[],"category_scores_codex":[0.001344517,0.0001340125,0.000370332,0.0000173374,0.0001887225,0.00008702136,0.0004856034,0.00003680649,0.00003514948],"category_scores_gemma":[0.0002634134,0.0000796243,0.00004995509,0.000139153,0.0003642281,0.0003708939,0.0001542397,0.0001545354,3.97242e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001435638,"about_ca_system_score_gemma":0.00004094129,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004634507,"about_ca_topic_score_gemma":0.00000313574,"domain_scores_codex":[0.9981596,0.0003607847,0.0006064463,0.000177507,0.000490875,0.0002047273],"domain_scores_gemma":[0.9976238,0.0005659491,0.001252789,0.0001951756,0.0002970845,0.00006519636],"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":[0.001101148,0.000210411,0.0004321514,0.00002831704,0.00001421366,6.786926e-7,0.002461458,0.03267973,0.9627785,0.0001466096,0.00008317694,0.0000636195],"study_design_scores_gemma":[0.000775765,0.0005344141,0.001698024,0.00006606791,0.00004761467,0.00002239798,0.0006123966,0.2410488,0.7488156,0.006248655,0.000003041475,0.0001272519],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9830633,0.00002133701,0.0161698,0.0003245233,0.0002441,0.000138838,0.000004422378,0.000008982933,0.00002471127],"genre_scores_gemma":[0.9980926,0.000001419832,0.001042068,0.00004268467,0.0008049865,0.000002309202,0.000001303384,0.000008165748,0.00000445651],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2139629,"threshold_uncertainty_score":0.3246985,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05107494095692278,"score_gpt":0.3710061537636172,"score_spread":0.3199312128066944,"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."}}