{"id":"W4387899421","doi":"10.1088/2632-2153/ad0652","title":"Neural network Gaussian processes as efficient models of potential energy surfaces for polyatomic molecules","year":2023,"lang":"en","type":"article","venue":"Machine Learning Science and Technology","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Gaussian process; Kernel (algebra); Polyatomic ion; Computer science; Gaussian; Function (biology); Gaussian function; Artificial neural network; Kriging; Generalization; Mathematical optimization; Algorithm; Artificial intelligence; Mathematics; Machine learning; Statistical physics; Computational chemistry; Physics; Chemistry; Molecule; 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":[0.002182305,0.000247959,0.000383049,0.0006961665,0.0009912289,0.0001987125,0.001093873,0.0001356753,0.00004248187],"category_scores_gemma":[0.001460562,0.0002081145,0.00003579791,0.003352852,0.002022037,0.000264961,0.000633678,0.000202699,0.00002176181],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003075768,"about_ca_system_score_gemma":0.0003259863,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003451997,"about_ca_topic_score_gemma":0.00002560868,"domain_scores_codex":[0.9970735,0.00009227776,0.0004065481,0.0008459272,0.0006512315,0.0009304842],"domain_scores_gemma":[0.9986869,0.0001573791,0.0003289359,0.0003404442,0.0003752697,0.0001111466],"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":[0.00002987348,0.0000180693,0.001149964,0.0000589122,0.000002322636,0.000006275689,0.0001071709,0.542061,0.4413962,0.01385768,0.000025609,0.001287007],"study_design_scores_gemma":[0.00028019,0.0003429501,0.0003829287,0.00004091223,0.0000129898,0.00005518336,0.0001448236,0.9051879,0.08070195,0.01225569,0.0003512697,0.0002432347],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9893541,0.0004327782,0.006841145,0.001812128,0.0004055792,0.0002154448,0.00001287675,0.0006805768,0.0002454021],"genre_scores_gemma":[0.9953429,0.00004547478,0.004182358,0.0000721628,0.00005024349,0.00004907236,0.00000725489,0.00002631355,0.0002242893],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3631269,"threshold_uncertainty_score":0.8486664,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00728694705171877,"score_gpt":0.2471794843492509,"score_spread":0.2398925372975321,"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."}}