{"id":"W2137743578","doi":"10.1109/34.845379","title":"Training hidden Markov models with multiple observations-a combinatorial method","year":2000,"lang":"en","type":"article","venue":"IEEE Transactions on Pattern Analysis and Machine Intelligence","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":150,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; Université Laval; SNC-Lavalin (Canada)","funders":"Natural Sciences and Engineering Research Council of Canada; Guangxi University; East China Institute of Technology","keywords":"Hidden Markov model; Computer science; Artificial intelligence; Markov chain; Lagrange multiplier; Generality; Handwriting; Independence (probability theory); Machine learning; Function (biology); Maximization; Pattern recognition (psychology); Algorithm; Mathematics; Mathematical optimization; Statistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003248538,0.0002392522,0.0003629132,0.0003897201,0.0002883438,0.0001968624,0.0004155757,0.00007424955,0.0007189052],"category_scores_gemma":[0.000004558187,0.0001979027,0.0002132384,0.001210967,0.00005009749,0.0004686539,0.000002754001,0.0002376214,0.00003072151],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000279866,"about_ca_system_score_gemma":0.00003005442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001131846,"about_ca_topic_score_gemma":0.001095866,"domain_scores_codex":[0.9982935,0.000166223,0.0003601315,0.0005752897,0.0003466858,0.0002581631],"domain_scores_gemma":[0.9988309,0.0003821563,0.00007775506,0.0004592969,0.00009014766,0.0001597675],"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":[0.00002689675,0.0001306334,0.0001301929,0.000004503069,0.0003601096,0.000009229564,0.0005573534,0.01038158,0.00003932579,0.0001693954,0.000003906467,0.9881869],"study_design_scores_gemma":[0.0002545113,0.0001218209,0.0006389078,0.00002502487,0.0003810089,0.00002597388,0.0001052207,0.9764352,0.0197282,0.001812085,0.0001273236,0.0003446957],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004376739,0.00001984084,0.9940422,0.0004556229,0.0001232009,0.0001418072,0.00003803956,0.0001361391,0.0006664431],"genre_scores_gemma":[0.8832632,0.0001413795,0.1156475,0.0005434065,0.00001969405,0.00004286923,0.000005837075,0.0000128648,0.0003232168],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9878422,"threshold_uncertainty_score":0.807024,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06005585691777636,"score_gpt":0.2821406138225669,"score_spread":0.2220847569047905,"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."}}