{"id":"W4220808348","doi":"10.1557/s43577-021-00183-4","title":"Machine learning as a tool to engineer microstructures: Morphological prediction of tannin-based colloids using Bayesian surrogate models","year":2022,"lang":"en","type":"article","venue":"MRS Bulletin","topic":"Pickering emulsions and particle stabilization","field":"Materials Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Australian Institute for Musculoskeletal Science; Academy of Finland; Finnish Center for Artificial Intelligence; Aalto-Yliopisto; European Commission","keywords":"Tannic acid; Suspension (topology); Materials science; Biological system; Particle (ecology); Nanotechnology; Colloid; Adhesion; Crystallization; Computer science; Artificial intelligence; Process engineering; Machine learning; Chemical engineering; Chemistry; Mathematics; Composite material; Organic chemistry; Engineering; Biology","routes":{"ca_aff":true,"ca_fund":false,"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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0004534164,0.0001139766,0.0001634093,0.00008585211,0.0002802357,0.00002976184,0.0001440174,0.00004039081,0.002870096],"category_scores_gemma":[0.0001450022,0.0001097469,0.00004668594,0.0002531665,0.00003761789,0.0000272952,0.0001164727,0.0001304142,0.00001731214],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000905228,"about_ca_system_score_gemma":0.00004273592,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002271972,"about_ca_topic_score_gemma":0.000001603149,"domain_scores_codex":[0.9987956,0.000166948,0.0002691843,0.0002636461,0.0002781811,0.000226474],"domain_scores_gemma":[0.999567,0.00005375672,0.00008166341,0.0001724471,0.0000582998,0.00006687038],"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":[0.00007119129,0.00002472306,0.0005425412,0.000007117072,0.000001370222,0.000005449982,0.0001672501,0.5025675,0.4964302,0.00004955516,0.00009215732,0.00004101531],"study_design_scores_gemma":[0.0009052714,0.0005562698,0.001085967,0.00003155013,0.00002427508,0.00003384349,0.0001781618,0.8236745,0.1568602,0.0002655363,0.01612765,0.000256719],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9570173,0.00004138894,0.04195043,0.0002655547,0.0001810209,0.0002289488,0.0001524427,0.00008793531,0.00007497078],"genre_scores_gemma":[0.989566,8.885177e-7,0.009942184,0.0001918222,0.0000216644,0.00003454171,0.00004367578,0.0000185602,0.0001806896],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3395699,"threshold_uncertainty_score":0.9980414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01769519477548507,"score_gpt":0.2360072396817332,"score_spread":0.2183120449062481,"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."}}