{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006054594,0.0009618683,0.0006720127,0.0003385351,0.0003290623,0.0006138646,0.001212837,0.0006017318,0.002293163],"category_scores_gemma":[0.00328596,0.0004836999,0.0004506797,0.0003664797,0.0005149915,0.001739106,0.00086,0.001309031,0.0006243218],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005268591,"about_ca_system_score_gemma":0.0009083538,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002182213,"about_ca_topic_score_gemma":0.006648745,"domain_scores_codex":[0.9996922,0.00004952403,0.00002737827,0.00008137769,0.00009081879,0.00005869912],"domain_scores_gemma":[0.9992667,0.0003607989,0.00009330386,0.0001180787,0.0001247713,0.00003648226],"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":[0.0004417735,0.0002283993,0.002192977,0.0003966461,0.0001721959,0.0005230357,0.000251959,0.4548068,0.08608972,0.02318506,0.006330169,0.4253813],"study_design_scores_gemma":[0.00001021761,0.0000412716,0.0001532217,0.000008996954,0.00001448837,0.0000431213,0.00001124884,0.9817197,0.01195374,0.005376545,0.0006610231,0.000006447278],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08768083,0.0003766195,0.9065208,0.0002277506,0.0001339377,0.00005314441,0.0001347865,0.002832881,0.002039224],"genre_scores_gemma":[0.750958,0.0002937535,0.2445818,0.0002393719,0.00006710167,0.0001214115,0.0003835976,0.0002632923,0.003091685],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002293163,"threshold_uncertainty_score":0.007671416,"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."}}