{"id":"W2953257883","doi":"10.14288/1.0223143","title":"Efficient algorithms for training the parameters of hidden Markov models using stochastic expectation maximization (EM) training and Viterbi training","year":2009,"lang":"en","type":"preprint","venue":"MDC Repository (Max-Delbrueck-Center for Molecular Medicine)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Viterbi algorithm; Computer science; Hidden Markov model; Forward algorithm; Iterative Viterbi decoding; Markov model; Markov chain; Soft output Viterbi algorithm; Expectation–maximization algorithm; Algorithm; Maximum-entropy Markov model; Machine learning; Training (meteorology); Artificial intelligence; Variable-order Markov model; Mathematics; Maximum likelihood; Decoding methods","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004019421,0.002231562,0.00155541,0.001477597,0.001004576,0.001554841,0.004211438,0.002788857,0.009247785],"category_scores_gemma":[0.02604236,0.001607044,0.001242754,0.001583975,0.001009372,0.002662904,0.002536939,0.004866339,0.004760115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001788785,"about_ca_system_score_gemma":0.003062282,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006002483,"about_ca_topic_score_gemma":0.008208998,"domain_scores_codex":[0.9978561,0.0009283285,0.000218757,0.0004835868,0.0003849545,0.0001283532],"domain_scores_gemma":[0.9899615,0.008004556,0.0004418803,0.0006867615,0.0007749175,0.0001304567],"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":[0.0002299357,0.0001601044,0.001442342,0.0003195238,0.0001568487,0.00009371424,0.0002472666,0.5010521,0.004379503,0.02272942,0.007280943,0.4619084],"study_design_scores_gemma":[0.00003334306,0.00002348056,0.0001395744,0.0000292377,0.00001652595,0.00004595066,0.00001980124,0.979185,0.002810957,0.01598859,0.001690114,0.00001732162],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001626857,0.0001573547,0.995837,0.000106284,0.00002684561,0.00005018181,0.00005992548,0.001729835,0.0004057786],"genre_scores_gemma":[0.0409283,0.0002622402,0.9556864,0.0001621832,0.00005566117,0.0004784902,0.000653131,0.0005724768,0.001201148],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009247785,"threshold_uncertainty_score":0.0309369,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05981510202793494,"score_gpt":0.2818756105327299,"score_spread":0.2220605085047949,"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."}}