{"id":"W2125113755","doi":"","title":"A Better Way to Pretrain Deep Boltzmann Machines","year":2012,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Generative Adversarial Networks and Image Synthesis","field":"Computer Science","cited_by":106,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"MNIST database; Boltzmann machine; Computer science; Restricted Boltzmann machine; Artificial intelligence; Deep belief network; Deep learning; Layer (electronics); Boltzmann constant; Generative grammar; Generative model; Pattern recognition (psychology); Machine learning; Algorithm","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":[0.0003875264,0.0001584826,0.0001602806,0.0001384535,0.0002395222,0.0008861122,0.0004036039,0.00005717866,0.000007168807],"category_scores_gemma":[0.0000592817,0.0001243911,0.0000422597,0.0003874329,0.00001709927,0.007782174,0.0001035684,0.00009913683,0.0001897702],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003787953,"about_ca_system_score_gemma":0.00001482981,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002546899,"about_ca_topic_score_gemma":0.000001580828,"domain_scores_codex":[0.9987293,0.00007484727,0.0003958625,0.0001245014,0.0002992432,0.0003761725],"domain_scores_gemma":[0.9992371,0.00002872373,0.0001781352,0.0002363306,0.0001563525,0.0001633443],"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":[0.00001080536,0.00003572588,0.001933107,0.000246032,0.0000186014,8.638789e-7,0.01469516,0.0267404,0.000852829,0.001867615,0.01154665,0.9420522],"study_design_scores_gemma":[0.0001294324,0.00003200135,0.002551678,0.00004806945,0.00000471633,0.00002585297,0.0001233496,0.9598376,0.0005000659,0.00002389307,0.03650178,0.0002215221],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009534892,0.0002997389,0.9837146,0.001151753,0.001789619,0.0003028664,0.000002410527,0.0001976318,0.003006513],"genre_scores_gemma":[0.9853023,8.180151e-7,0.01199801,0.0018945,0.0006150115,0.0000525731,0.000007027254,0.000006757988,0.000122967],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9757674,"threshold_uncertainty_score":0.85448,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0145427736631149,"score_gpt":0.2387916199599926,"score_spread":0.2242488462968777,"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."}}