{"id":"W3105027165","doi":"","title":"Regularized linear autoencoders recover the principal components, eventually","year":2020,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Regularization (linguistics); Feature learning; Computer science; Simple (philosophy); Representation (politics); Principal component analysis; Gradient descent; Mathematical optimization; Mathematics; Artificial intelligence; Artificial neural network; Algorithm; Applied mathematics","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":[0.0002083175,0.0001411597,0.0001498585,0.00003403312,0.0004098012,0.0007425165,0.0008516099,0.00005408696,0.00000319592],"category_scores_gemma":[0.00003499799,0.0000955298,0.00005734236,0.0005143154,0.0000384713,0.0026009,0.0001545349,0.0002116022,0.0001066931],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002846326,"about_ca_system_score_gemma":0.00006410573,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001690664,"about_ca_topic_score_gemma":5.528444e-7,"domain_scores_codex":[0.9986195,0.00006604307,0.0005218366,0.0001759102,0.000389336,0.0002273746],"domain_scores_gemma":[0.9990398,0.00004507902,0.0003563776,0.0002947143,0.0001513473,0.0001126524],"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":[0.0001112878,0.0001093598,0.0004414486,0.001055272,0.00008046443,0.000008080056,0.02084807,0.6489385,0.003289429,0.08619871,0.04680034,0.1921191],"study_design_scores_gemma":[0.0002155715,0.00002650603,0.000179767,0.00002903874,0.000003947135,0.00001693843,0.00008722395,0.9126045,0.00004968538,0.00005066002,0.08662254,0.0001136341],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01475653,0.0001414765,0.9571496,0.02421715,0.0006490988,0.0008443351,0.000006744378,0.000675436,0.00155962],"genre_scores_gemma":[0.9912022,0.000006335241,0.003754637,0.004546544,0.0002411442,0.00007666882,0.0000273925,0.000007778746,0.000137248],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9764457,"threshold_uncertainty_score":0.7160103,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03675184295725883,"score_gpt":0.2432656193508974,"score_spread":0.2065137763936386,"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."}}