{"id":"W2954365962","doi":"10.48550/arxiv.1806.05819","title":"BubbleRank: Safe Online Learning to Re-Rank via Implicit Click Feedback","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Learning to rank; Computer science; Ranking (information retrieval); Regret; Rank (graph theory); Online learning; Relevance (law); Quality (philosophy); Base (topology); Information retrieval; Machine learning; Artificial intelligence; World Wide Web; Mathematics","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.002433266,0.0006903532,0.00107954,0.001485526,0.0006291503,0.0004001396,0.004492973,0.0006671713,0.002819939],"category_scores_gemma":[0.002554445,0.0006842391,0.0005331304,0.003179637,0.00046861,0.0005356717,0.005773919,0.002210201,0.007400688],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006471429,"about_ca_system_score_gemma":0.0003523028,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003948399,"about_ca_topic_score_gemma":0.0004946473,"domain_scores_codex":[0.9928712,0.0007176884,0.0008993752,0.003145311,0.001147203,0.001219253],"domain_scores_gemma":[0.9924526,0.001458026,0.0006412917,0.002782941,0.00179962,0.0008654997],"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":[0.0008163798,0.0003211536,0.01049521,0.00005111969,0.0002080528,0.0006524205,0.0009427807,0.9442927,0.0003159395,0.0004859439,0.009590076,0.03182817],"study_design_scores_gemma":[0.003532301,0.001180138,0.03852364,0.0004157509,0.0001959761,0.00003714003,0.003787641,0.6107159,0.0007814661,0.1850982,0.1524081,0.003323732],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5069759,0.00005415,0.4865701,0.00045632,0.0009170268,0.000777624,0.0001308439,0.000219695,0.003898357],"genre_scores_gemma":[0.9473605,0.0001621295,0.003728879,0.0002557384,0.0007854779,0.000002727994,0.00007608208,0.00008717374,0.04754129],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4828412,"threshold_uncertainty_score":0.9995609,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.213937048042641,"score_gpt":0.332489253193239,"score_spread":0.118552205150598,"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."}}