{"id":"W2109569768","doi":"10.1109/iscas.2009.5117696","title":"Second order cone programming (SOCP) relaxations for large margin HMMs in speech recognition","year":2009,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Computer science; Hidden Markov model; Margin (machine learning); Second-order cone programming; Speech recognition; Convex optimization; String (physics); Pattern recognition (psychology); Regular polygon; Artificial intelligence; Algorithm; Mathematical optimization; Mathematics; Machine learning","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":[0.0003083837,0.0001010926,0.0001199912,0.0001344765,0.0001262007,0.0002072961,0.0002315849,0.00006700507,0.0001248085],"category_scores_gemma":[0.0001143892,0.00009562387,0.00003643883,0.0006156295,0.000008589008,0.000757509,0.00003102408,0.0001066446,0.00004583895],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003644334,"about_ca_system_score_gemma":0.00006645556,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000423698,"about_ca_topic_score_gemma":0.0001719204,"domain_scores_codex":[0.9989734,0.00001776801,0.0002226862,0.0003039455,0.000122373,0.0003598476],"domain_scores_gemma":[0.9994898,0.00005729212,0.00008179894,0.0001748653,0.0001387001,0.00005755147],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00000681506,0.0001843957,0.0003746049,0.00001962298,0.000004280571,0.000008467954,0.0002898726,0.000001768739,0.002392732,0.001842235,0.001867344,0.9930078],"study_design_scores_gemma":[0.008322354,0.0009263185,0.0119213,0.0003556925,0.00002440511,0.0001361798,0.000677183,0.02342446,0.6241564,0.1545379,0.1738524,0.001665471],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06893723,0.0001203375,0.9162294,0.002989468,0.000161967,0.0006108242,0.000006051363,0.0002917185,0.01065303],"genre_scores_gemma":[0.1934933,0.000005024071,0.8028299,0.001427067,0.00006797791,0.00004191738,0.00002768141,0.000007190926,0.002100025],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9913424,"threshold_uncertainty_score":0.3899428,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0214147609913022,"score_gpt":0.2713393307364871,"score_spread":0.2499245697451849,"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."}}