{"id":"W4224231654","doi":"10.1111/risa.13930","title":"Subjective machines: Probabilistic risk assessment based on deep learning of soft information","year":2022,"lang":"en","type":"article","venue":"Risk Analysis","topic":"Risk and Safety Analysis","field":"Decision Sciences","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Machine learning; Probabilistic logic; Artificial intelligence; Computer science; Heuristics; Gradient boosting; Risk assessment; Process (computing); Random forest","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":[0.007523582,0.0002081246,0.0007475452,0.002345511,0.001105437,0.0001602917,0.0007152298,0.00005405889,0.002886374],"category_scores_gemma":[0.005632661,0.0001607995,0.001062579,0.008840877,0.00008246383,0.0004156994,0.000177158,0.0006535344,0.0000898791],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002404143,"about_ca_system_score_gemma":0.0001282896,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002073437,"about_ca_topic_score_gemma":0.0005407098,"domain_scores_codex":[0.992591,0.002554157,0.001221281,0.0004782547,0.00290132,0.0002539346],"domain_scores_gemma":[0.993562,0.003084641,0.001815205,0.0008564115,0.0005758808,0.0001058884],"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":[0.00005566646,0.00007314421,0.3678688,0.000001562694,0.0003769893,7.743586e-7,0.0004997362,0.5694123,9.586325e-7,0.0001195163,0.0000250244,0.06156551],"study_design_scores_gemma":[0.0002399865,0.0001470816,0.2781967,0.000001144958,0.001715511,1.952373e-7,0.00200832,0.7112542,0.000003514901,0.005416689,0.0008883306,0.0001284025],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3838215,0.00006186898,0.6114032,0.0002537072,0.00007905236,0.0002216236,0.0002288571,0.00005425039,0.003875974],"genre_scores_gemma":[0.9974927,0.00007359244,0.001998018,0.00006001836,0.00001695821,0.0000516525,0.0001317935,0.000007865267,0.0001674406],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6136712,"threshold_uncertainty_score":0.9980251,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01601607476966815,"score_gpt":0.3191089711325679,"score_spread":0.3030928963628997,"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."}}