{"id":"W4415233468","doi":"10.48550/arxiv.2507.18242","title":"Boosting Revisited: Benchmarking and Advancing LP-Based Ensemble Methods","year":2025,"lang":"en","type":"preprint","venue":"University of Twente Research Information","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds de recherche du Québec – Nature et technologies; Alliance de recherche numérique du Canada; TKI DINALOG; Compute Canada","keywords":"Boosting (machine learning); Heuristics; Ensemble learning; Benchmarking; Decision tree; Hyperparameter; Gradient boosting; Margin (machine learning)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004601597,0.0001118934,0.000197676,0.0009658439,0.0003949122,0.0001968772,0.0008515812,0.000142846,0.000008538018],"category_scores_gemma":[0.0009428277,0.0001346495,0.00005634386,0.0004745258,0.00007481933,0.001199639,0.002072047,0.000820045,0.000005644852],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001799175,"about_ca_system_score_gemma":0.000403642,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006081312,"about_ca_topic_score_gemma":0.00001356224,"domain_scores_codex":[0.9981208,0.0006352759,0.0002473573,0.0002628909,0.0004759264,0.0002577598],"domain_scores_gemma":[0.9977596,0.0006230876,0.0003368737,0.0006134113,0.0005785464,0.000088455],"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":[0.00003008161,0.00001731964,0.001261774,0.001700458,0.00002575669,0.00000184412,0.00185287,0.004737356,0.0001586595,0.01061762,0.002273735,0.9773225],"study_design_scores_gemma":[0.0002562624,0.0000415037,0.002326428,0.0008800635,0.00001021078,8.966969e-7,0.0003836936,0.9268838,0.0001936279,0.0006176956,0.06827694,0.0001288332],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0009211536,0.0001316563,0.9832326,0.001053918,0.0001056547,0.0002743975,0.00002270477,0.00008071645,0.01417725],"genre_scores_gemma":[0.05165569,0.0004937675,0.9469551,0.0000716284,0.00002989867,0.000001091772,0.0004409971,0.00000342499,0.0003483403],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9771937,"threshold_uncertainty_score":0.5490847,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0482439719890473,"score_gpt":0.3789084719229865,"score_spread":0.3306644999339393,"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."}}