{"id":"W3124175555","doi":"10.15353/jcvis.v6i1.3538","title":"Where Should We Begin? A Low-Level Exploration of Weight Initialization Impact on Quantized Behaviour of Deep Neural Networks","year":2021,"lang":"en","type":"article","venue":"Journal of Computational Vision and Imaging Systems","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Initialization; Quantization (signal processing); Convolutional neural network; Computer science; Inference; Artificial intelligence; Deep neural networks; Artificial neural network; Deep learning; Algorithm","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"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":[0.0003279246,0.0001625167,0.0003935052,0.0002268965,0.00009534469,0.0001204218,0.0002430071,0.00004548563,0.000004641108],"category_scores_gemma":[0.00003972973,0.0001289663,0.0001412153,0.0005377472,0.00005088668,0.001177852,0.00007024233,0.0001839741,6.19091e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004474857,"about_ca_system_score_gemma":0.00008285038,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004797565,"about_ca_topic_score_gemma":0.000001185196,"domain_scores_codex":[0.9979488,0.0002263687,0.0009131189,0.0001838773,0.0005886158,0.0001392377],"domain_scores_gemma":[0.9971213,0.0004002429,0.001116205,0.0001886243,0.001072257,0.0001013087],"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":[0.00005537591,0.0001484374,0.00171009,0.0000540955,0.00002831084,0.00001532912,0.0003288209,0.9713766,0.0007786673,0.007722313,0.000352994,0.01742895],"study_design_scores_gemma":[0.0007868396,0.0001598457,0.007458428,0.0004100514,0.00001956568,0.0002257395,0.0001238432,0.987554,0.000239552,0.002867921,0.00004111034,0.0001131477],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04343393,0.00209713,0.9524664,0.001339647,0.0004745024,0.0001444364,0.000006704774,0.00001671564,0.00002048365],"genre_scores_gemma":[0.9877394,0.0001953704,0.01180835,0.00007536219,0.0001432412,0.000003304529,0.00001540058,0.00001306552,0.000006456048],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9443055,"threshold_uncertainty_score":0.5259094,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04322401672907961,"score_gpt":0.3389066658185412,"score_spread":0.2956826490894616,"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."}}