{"id":"W2917507868","doi":"10.1101/564476","title":"Dynamic compression and expansion in a classifying recurrent network","year":2019,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Neural dynamics and brain function","field":"Neuroscience","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Fonds de recherche du Québec – Nature et technologies; Fonds de Recherche du Québec - Santé; Natural Sciences and Engineering Research Council of Canada; National Science Foundation","keywords":"Curse of dimensionality; Computer science; Dimensionality reduction; Recurrent neural network; Artificial intelligence; Stochastic gradient descent; Gradient descent; Generalization; Artificial neural network; Task (project management); Chaotic; Machine learning; Mathematics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004315957,0.000494247,0.000531479,0.0002667268,0.0001771605,0.0002374956,0.0003797604,0.0004512648,0.00001305287],"category_scores_gemma":[0.0002360409,0.0004880879,0.00009288193,0.0004752847,0.00009665833,0.0001959267,0.0008864757,0.001191718,0.00003251498],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002610843,"about_ca_system_score_gemma":0.0001673935,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001851379,"about_ca_topic_score_gemma":0.000005759099,"domain_scores_codex":[0.996774,0.0002989557,0.0005137305,0.001420163,0.0003920675,0.0006010542],"domain_scores_gemma":[0.9983197,0.0002056567,0.0003896226,0.0008376585,0.00007536837,0.0001719619],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007629144,0.00009205718,0.004446572,0.0003420954,0.000006128498,0.00004901529,0.000006406035,0.00324755,0.9912552,0.0003245868,0.0001210653,0.00003304081],"study_design_scores_gemma":[0.001697581,0.0002451193,0.2754428,0.004622622,0.00006549154,1.32694e-7,0.000003785967,0.6365226,0.07579543,0.00005613258,0.003568946,0.001979341],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9934235,0.000702189,0.0008216813,0.0002250824,0.003638028,0.0009043835,0.00005253948,0.0002212903,0.00001126188],"genre_scores_gemma":[0.997569,0.001074465,0.0006859308,0.0003404102,0.0001467972,0.00008474847,3.125278e-7,0.00009074589,0.00000753578],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9154598,"threshold_uncertainty_score":0.9997571,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0241039545711107,"score_gpt":0.2416053100240541,"score_spread":0.2175013554529434,"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."}}