{"id":"W3134545286","doi":"10.1038/s41467-021-21696-1","title":"Predictive learning as a network mechanism for extracting low-dimensional latent space representations","year":2021,"lang":"en","type":"article","venue":"Nature Communications","topic":"Neural dynamics and brain function","field":"Neuroscience","cited_by":84,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Natural Sciences and Engineering Research Council of Canada; University of Oregon; Fonds de recherche du Québec – Nature et technologies; University of Washington; DeepMind; Harvard University; National Science Foundation","keywords":"Computer science; Artificial intelligence; Artificial neural network; Latent variable; Curse of dimensionality; Machine learning; Latent variable model; Task (project management); Recurrent neural network; Space (punctuation); Focus (optics); Sequence learning","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":[0.0001880237,0.0001103768,0.0001136793,0.00004484736,0.001191571,0.00008360694,0.0003527522,0.0001624747,0.00002819561],"category_scores_gemma":[0.003600488,0.0001135336,0.0001051664,0.0004505907,0.00005783837,0.0001798998,0.0003288843,0.00105535,0.00001954955],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005328789,"about_ca_system_score_gemma":0.0001006715,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007455617,"about_ca_topic_score_gemma":0.00006025305,"domain_scores_codex":[0.9987454,0.0002778849,0.0001951267,0.0003494885,0.0002114402,0.0002205983],"domain_scores_gemma":[0.9965325,0.002149204,0.0001535045,0.0007724203,0.0003260192,0.00006633549],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004213752,0.0001734972,0.0002963211,0.000008742627,0.00002223638,0.00000858388,0.0001745472,0.01058134,0.2965143,0.6888148,0.002815875,0.0005476493],"study_design_scores_gemma":[0.001222639,0.000242938,0.003534913,0.0001991324,0.0001419388,0.0002079963,0.0005597225,0.6998397,0.1091357,0.1531263,0.03116665,0.0006223663],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6400379,0.004383938,0.05645392,0.2208274,0.008968812,0.005341804,0.0003926719,0.002015792,0.06157784],"genre_scores_gemma":[0.9856467,0.00020939,0.008938347,0.002225115,0.0001262171,0.0001235622,0.0001378906,0.00002351833,0.002569224],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6892584,"threshold_uncertainty_score":0.9164723,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03485729945360345,"score_gpt":0.3220740127034992,"score_spread":0.2872167132498958,"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."}}