{"id":"W4285601419","doi":"10.24963/ijcai.2022/738","title":"Homeomorphic-Invariance of EM: Non-Asymptotic Convergence in KL Divergence for Exponential Families via Mirror Descent (Extended Abstract)","year":2022,"lang":"en","type":"article","venue":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; University of British Columbia","funders":"","keywords":"Bregman divergence; Divergence (linguistics); Kullback–Leibler divergence; Applied mathematics; Mathematics; Exponential family; Gradient descent; Gaussian; Convergence (economics); Invariant (physics); Exponential function; Probabilistic logic; Parametrization (atmospheric modeling); Expectation–maximization algorithm; Mathematical optimization; Computer science; Mathematical analysis; Artificial neural network; Artificial intelligence; Physics; Statistics; Maximum likelihood","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"],"consensus_categories":[],"category_scores_codex":[0.001371909,0.0003352951,0.0005412344,0.0005204214,0.0002981517,0.000122412,0.003307719,0.00009507684,0.0002318754],"category_scores_gemma":[0.0003255214,0.0002985718,0.0003589771,0.001075079,0.0002685801,0.0004721467,0.001032056,0.0004816988,0.00000871383],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001889295,"about_ca_system_score_gemma":0.0001812085,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000300499,"about_ca_topic_score_gemma":0.000128374,"domain_scores_codex":[0.99638,0.00004099526,0.00123041,0.0008024734,0.001094494,0.0004516105],"domain_scores_gemma":[0.9973931,0.0002020288,0.0009914219,0.0003934243,0.0009160064,0.0001039839],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004654154,0.0008653856,0.001267119,0.0001721063,0.000168096,0.000003922851,0.004866882,0.001239146,0.0818725,0.8920283,0.00008321132,0.01696794],"study_design_scores_gemma":[0.0002282258,0.0004465786,0.0119664,0.0003234496,0.00006343002,0.000014561,0.000969956,0.333966,0.3465595,0.3048327,0.00008970201,0.0005395171],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1602877,0.00002904244,0.8333516,0.002216195,0.002296193,0.0009441171,0.00007860959,0.0000451362,0.0007513872],"genre_scores_gemma":[0.9579263,0.00008281176,0.04142587,0.0001890548,0.00004492021,0.000211265,0.0000041975,0.00001688679,0.00009871841],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7976385,"threshold_uncertainty_score":0.9999467,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06029236595116157,"score_gpt":0.2829675120279982,"score_spread":0.2226751460768367,"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."}}