{"id":"W2107999479","doi":"10.1109/icnn.1996.549206","title":"The effects of reduced precision bit lengths on feedforward neural networks for speech recognition","year":2002,"lang":"en","type":"article","venue":"Proceedings of International Conference on Neural Networks (ICNN'96)","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Technical University of Nova Scotia","funders":"","keywords":"Artificial neural network; Computer science; Feedforward neural network; Feed forward; Fixed-point arithmetic; Speech recognition; Fixed point; Floating point; Set (abstract data type); Arithmetic; A priori and a posteriori; Time delay neural network; Precision and recall; Independence (probability theory); Point (geometry); Algorithm; Artificial intelligence; Pattern recognition (psychology); Mathematics; Statistics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003927161,0.0004122132,0.0004186899,0.0001906561,0.0003505874,0.0003990553,0.001956803,0.0001955714,0.00002720614],"category_scores_gemma":[0.00031604,0.0003156226,0.0002930262,0.0005817855,0.0001678238,0.0006780433,0.0002785981,0.0005704369,0.000007960486],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006594498,"about_ca_system_score_gemma":0.00001228862,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008289609,"about_ca_topic_score_gemma":0.000003121584,"domain_scores_codex":[0.9970487,0.00003373013,0.000835528,0.00078748,0.0007234413,0.0005710814],"domain_scores_gemma":[0.9962342,0.001235758,0.000835359,0.0003350098,0.001198051,0.0001616444],"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":[0.000544786,0.0004518841,0.0002460192,0.00008916384,0.0001275157,0.000003733999,0.0001461357,0.009924307,0.006459362,0.08797117,0.02091725,0.8731187],"study_design_scores_gemma":[0.0007723301,0.001055053,0.0003404105,0.0003452215,0.00002565488,0.00001784933,0.00002289811,0.9811982,0.009977424,0.005042227,0.0008828292,0.0003198827],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9342815,0.0006896866,0.02414519,0.01838286,0.006327165,0.004464705,0.00004655505,0.0004797663,0.01118254],"genre_scores_gemma":[0.9954886,0.001029721,0.001511273,0.0004818258,0.0007530771,0.0002662781,0.00002369954,0.00003568682,0.0004097906],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9712739,"threshold_uncertainty_score":0.9999296,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03712323545250512,"score_gpt":0.2696142770525777,"score_spread":0.2324910416000726,"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."}}