{"id":"W4388567027","doi":"10.1038/s41467-023-42992-y","title":"Exploiting redundancy in large materials datasets for efficient machine learning with less data","year":2023,"lang":"en","type":"article","venue":"Nature Communications","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":98,"is_retracted":false,"has_abstract":true,"ca_institutions":"Schwartz/Reisman Emergency Medicine Institute; Vector Institute; Natural Resources Canada; University of Toronto; University of New Brunswick","funders":"Office of Energy Research and Development; Natural Resources Canada; Canada First Research Excellence Fund; University of Toronto; National Institute of Standards and Technology; Alliance de recherche numérique du Canada; Western Canada Research Grid","keywords":"Computer science; Redundancy (engineering); Robustness (evolution); Machine learning; Training set; Artificial intelligence; Data mining","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":[0.004692326,0.0002071865,0.0003079337,0.0002271076,0.0008288486,0.000300708,0.004308906,0.0001532529,0.0001194526],"category_scores_gemma":[0.001975022,0.000176697,0.00001957473,0.0008306291,0.0001741679,0.0003424858,0.002818193,0.0006252956,0.0001396112],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005419812,"about_ca_system_score_gemma":0.00009391242,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001610708,"about_ca_topic_score_gemma":0.001113451,"domain_scores_codex":[0.9973787,0.0005761287,0.0004640256,0.000641323,0.0003629997,0.0005767913],"domain_scores_gemma":[0.995089,0.0008190173,0.0002899966,0.003634125,0.0000949505,0.00007292727],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001733771,0.000370192,0.008785512,0.0002386929,0.00001608567,0.00001704754,0.00150605,0.01506478,0.9528914,0.01560497,0.004645503,0.0006863716],"study_design_scores_gemma":[0.004690066,0.0003150549,0.03322636,0.001156875,0.0001130298,0.00008203434,0.00267119,0.6211612,0.09174378,0.0006358385,0.242184,0.002020487],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9823859,0.0008290478,0.002491546,0.003874543,0.0007120064,0.001081064,0.007597078,0.0007053958,0.0003234741],"genre_scores_gemma":[0.9575405,0.00008816753,0.03143743,0.0001351196,0.00006777794,0.0001805758,0.01042143,0.00004473905,0.0000842516],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8611476,"threshold_uncertainty_score":0.8007091,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0605409308656181,"score_gpt":0.3639573872727498,"score_spread":0.3034164564071317,"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."}}