{"id":"W4403458004","doi":"10.1016/j.ins.2024.121564","title":"Tree smoothing: Post-hoc regularization of tree ensembles for interpretable machine learning","year":2024,"lang":"en","type":"article","venue":"Information Sciences","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Austrian Science Fund","keywords":"Artificial intelligence; Regularization (linguistics); Smoothing; Tree (set theory); Computer science; Machine learning; Post hoc; Pattern recognition (psychology); Mathematics; Combinatorics; Medicine","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":[],"consensus_categories":[],"category_scores_codex":[0.001442482,0.000114305,0.0001373704,0.0005009473,0.0003170442,0.0009206681,0.0008297058,0.00005471047,0.00002200419],"category_scores_gemma":[0.000512339,0.00009538124,0.00007802046,0.001202563,0.0001478054,0.007793472,0.0001383225,0.00009503733,0.00006358843],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003581252,"about_ca_system_score_gemma":0.000195142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008970698,"about_ca_topic_score_gemma":0.00006764442,"domain_scores_codex":[0.9985144,0.00004431657,0.0005249093,0.000212069,0.0004470815,0.0002572522],"domain_scores_gemma":[0.998924,0.0003305523,0.0001855656,0.000201614,0.0003090443,0.00004924246],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001933633,0.00001550182,0.000104377,0.0001367814,0.00001189293,7.756143e-7,0.01607144,0.008874127,0.005373507,0.3946871,0.0005217534,0.5741834],"study_design_scores_gemma":[0.00004931432,0.0003382511,0.00003716072,0.0000834915,0.000004326942,0.000006474808,0.001173679,0.9287112,0.03856662,0.01212868,0.0187794,0.0001213684],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008243128,0.0002823919,0.9758139,0.001775389,0.0005496253,0.0002482359,0.000006437833,0.0002509225,0.01283001],"genre_scores_gemma":[0.9639286,0.00003075684,0.03515857,0.0003058604,0.0000325002,0.00002722718,0.00001685286,0.00000507911,0.0004945116],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9556855,"threshold_uncertainty_score":0.8878024,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02549530159241103,"score_gpt":0.2852611363255154,"score_spread":0.2597658347331044,"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."}}