{"id":"W4392704564","doi":"10.29363/nanoge.matsus.2024.151","title":"Prediction Robustness and Data Redundancy in Machine Learning for Materials Science","year":2023,"lang":"en","type":"article","venue":"","topic":"Mineral Processing and Grinding","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Natural Resources Canada; University of Toronto","funders":"","keywords":"Robustness (evolution); Computer science; Redundancy (engineering); Machine learning; Artificial intelligence; Data mining","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00056599,0.00004750772,0.00006331571,0.0001344717,0.00009104532,0.00007999293,0.0001141187,0.0000200515,0.00001029976],"category_scores_gemma":[0.00009338519,0.00004304558,0.000002441691,0.0003142375,0.00002434719,0.0003054166,0.00007322484,0.00004467911,0.000002139515],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001274336,"about_ca_system_score_gemma":0.000008116231,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002043037,"about_ca_topic_score_gemma":0.00001473168,"domain_scores_codex":[0.9995355,0.000004002682,0.00009740516,0.0001531984,0.00005901643,0.000150889],"domain_scores_gemma":[0.9998403,0.00002023341,0.000008381321,0.00009945987,0.000009056711,0.00002257273],"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":[0.00001115308,0.000006374466,0.005658766,0.0005698554,0.000006002585,0.000002928578,0.0003520726,0.6355557,0.3371912,0.0002443875,0.001228838,0.01917272],"study_design_scores_gemma":[0.0001434573,0.000007609379,0.003049624,0.0000448485,0.000001886667,0.000002517895,0.00004288292,0.9907676,0.005420439,0.00004598242,0.0004170261,0.00005612672],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9937857,0.0000668529,0.005026308,0.00004108815,0.0002385134,0.00007100186,0.00002082251,0.0003503851,0.0003993228],"genre_scores_gemma":[0.9982843,0.00004100248,0.001039255,0.000002155954,0.00005162002,0.000008002443,0.00008137665,0.00001036642,0.0004818688],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3552119,"threshold_uncertainty_score":0.1755348,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05732945414698173,"score_gpt":0.2881150647884438,"score_spread":0.2307856106414621,"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."}}