{"id":"W2901663942","doi":"10.23919/date.2019.8714765","title":"Learning to Skip Ineffectual Recurrent Computations in LSTMs","year":2019,"lang":"en","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Pruning; Computation; Inference; Recurrent neural network; Speedup; Set (abstract data type); Process (computing); Computational complexity theory; Parallel computing; Algorithm; Artificial intelligence; Artificial neural network; Programming language","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00009652084,0.00007785454,0.00009396283,0.0001106374,0.00004795456,0.00004211373,0.0003962701,0.00002043609,0.0000269422],"category_scores_gemma":[0.00002804134,0.00007390203,0.00002106456,0.0008587191,0.000007257151,0.0002063643,0.0002225872,0.00016139,0.001202597],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004080666,"about_ca_system_score_gemma":0.00001556044,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006968889,"about_ca_topic_score_gemma":0.00002235313,"domain_scores_codex":[0.999184,0.00004506639,0.0001461823,0.00030668,0.0001204819,0.0001976087],"domain_scores_gemma":[0.999339,0.0002570767,0.00003033969,0.0002710985,0.00003039964,0.00007207505],"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":[0.000005103453,0.00008643055,0.005646849,0.000007609583,0.000003443266,0.00000232683,0.001471534,0.5806128,0.001189709,0.1401682,0.001560898,0.2692452],"study_design_scores_gemma":[0.0004208425,0.000295819,0.02090353,0.00003207468,0.000001070619,0.000006942331,0.0000677073,0.9443473,0.0004716111,0.005046874,0.02811526,0.0002909434],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2167758,0.00001315603,0.7767251,0.001829546,0.0001567373,0.0004207267,2.193173e-7,0.000203351,0.003875375],"genre_scores_gemma":[0.9113618,0.000003384336,0.08740254,0.0004727429,0.00002358005,0.0000496727,0.000002713748,0.000005814766,0.0006778229],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.694586,"threshold_uncertainty_score":0.9995751,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01319806794252159,"score_gpt":0.2819775062096999,"score_spread":0.2687794382671783,"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."}}