{"id":"W4382537648","doi":"10.1007/s42979-023-01920-z","title":"Training Integer-Only Deep Recurrent Neural Networks","year":2023,"lang":"en","type":"article","venue":"SN Computer Science","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University; Polytechnique Montréal; Huawei Technologies (Canada)","funders":"Huawei Technologies","keywords":"Computer science; Normalization (sociology); Recurrent neural network; Computation; Quantization (signal processing); Edge device; Artificial neural network; Algorithm; Piecewise linear function; Floating point; Artificial intelligence; 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":[0.001338227,0.0001993631,0.0002003717,0.0003921039,0.0004480891,0.0007095729,0.003421778,0.00004973849,0.000004995012],"category_scores_gemma":[0.00004979403,0.0001841793,0.0000817688,0.003233246,0.0002402687,0.001183135,0.001751476,0.0003057089,0.00008502159],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007772379,"about_ca_system_score_gemma":0.000217712,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009545816,"about_ca_topic_score_gemma":0.000008282803,"domain_scores_codex":[0.9969464,0.00006154312,0.000329793,0.001022127,0.0007089821,0.000931157],"domain_scores_gemma":[0.9983963,0.0001330814,0.00009373554,0.0009763087,0.0001360785,0.0002644988],"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":[6.323382e-7,0.000008774226,0.0001827438,0.00000329917,0.000002083254,0.00002737113,0.00183874,0.1038977,0.00002420454,0.01547396,0.0001768699,0.8783637],"study_design_scores_gemma":[0.0001100409,0.0000604128,0.00161252,0.00002548042,0.000001300825,0.00005820789,0.00002495051,0.9957142,0.00002230721,0.001494085,0.0006530855,0.0002233657],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02683409,0.00005630785,0.9652047,0.0009983184,0.005415349,0.0001300155,3.175845e-7,0.0008132093,0.0005476546],"genre_scores_gemma":[0.8845843,0.000007346719,0.1141807,0.0006833582,0.0004864674,0.000008101011,0.000001171029,0.000009225435,0.00003934731],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8918166,"threshold_uncertainty_score":0.7510614,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06051638147265736,"score_gpt":0.2882258716313546,"score_spread":0.2277094901586972,"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."}}