{"id":"W2895302585","doi":"10.1007/s11063-018-9923-4","title":"Low-Dimensional Perturb-and-MAP Approach for Learning Restricted Boltzmann Machines","year":2018,"lang":"en","type":"article","venue":"Neural Processing Letters","topic":"Generative Adversarial Networks and Image Synthesis","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; University of British Columbia","funders":"","keywords":"Maximum a posteriori estimation; Boltzmann machine; Algorithm; Mathematics; Gradient descent; Restricted Boltzmann machine; Boltzmann distribution; Function (biology); Divergence (linguistics); Computer science; Energy (signal processing); Mathematical optimization; Artificial intelligence; Statistical physics; Deep learning; Artificial neural network; 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":[],"consensus_categories":[],"category_scores_codex":[0.0001399604,0.0001904548,0.0001805593,0.00008093905,0.0006659031,0.0003520495,0.0003346561,0.00005018862,0.000003122047],"category_scores_gemma":[0.00008294584,0.0001535568,0.00005726673,0.0002210255,0.0001567032,0.0005970082,0.0001104161,0.0001743829,0.000003571999],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001615069,"about_ca_system_score_gemma":0.00003532767,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001192068,"about_ca_topic_score_gemma":9.151269e-7,"domain_scores_codex":[0.9986555,0.00008153837,0.0001823615,0.0005222741,0.0001996639,0.0003586933],"domain_scores_gemma":[0.9994193,0.00009963434,0.0001081858,0.0001635718,0.0001249498,0.0000843065],"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":[0.0001588198,0.0001348436,0.00085879,0.0002818118,0.00007083984,0.00001845345,0.003231812,0.0397666,0.3823422,0.0003596821,0.02710648,0.5456696],"study_design_scores_gemma":[0.0002999683,0.00009128624,0.0004941121,0.0000257844,0.00001081637,0.0000128318,0.0000138341,0.99404,0.002562032,0.00009924268,0.002134293,0.0002157688],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06569704,0.0001949889,0.9258681,0.007327552,0.0003551211,0.0002059335,0.000001079195,0.0001875563,0.0001626428],"genre_scores_gemma":[0.8487504,0.000001370792,0.1462138,0.004084757,0.0007310225,0.00002056681,0.000008686979,0.0000173023,0.0001720818],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9542734,"threshold_uncertainty_score":0.6261864,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01484261004209037,"score_gpt":0.2351871291408649,"score_spread":0.2203445190987746,"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."}}