{"id":"W4417298805","doi":"10.48550/arxiv.2505.14371","title":"Layer-wise Quantization for Quantized Optimistic Dual Averaging","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Army Research Office; Centers for Disease Control and Prevention; National Supercomputing Centre Singapore; Norges Forskningsråd; Institute for Catastrophic Loss Reduction; Hasler Stiftung; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; Centro Svizzero di Calcolo Scientifico; National Science Foundation","keywords":"Quantization (signal processing); Speedup; Monotone polygon; Convergence (economics); Representation (politics); Artificial neural network; Dual (grammatical number)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004907617,0.0004445241,0.0005298128,0.0004700608,0.0002425292,0.0002697186,0.001353077,0.0003355186,0.00001600781],"category_scores_gemma":[0.0006871547,0.0004892352,0.0002386742,0.0003960318,0.00006362292,0.0002514715,0.001410228,0.0003986267,0.00002001794],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001592786,"about_ca_system_score_gemma":0.0003187567,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003426897,"about_ca_topic_score_gemma":0.000003253564,"domain_scores_codex":[0.9972541,0.00009850886,0.0006753455,0.001163539,0.0003421148,0.0004663785],"domain_scores_gemma":[0.9975092,0.0004427352,0.0004419913,0.001135279,0.0003544542,0.0001162709],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001680733,0.0007353848,0.03669703,0.002128542,0.0006111372,0.00004395923,0.003151964,0.3331323,0.00149397,0.59797,0.01507035,0.008797274],"study_design_scores_gemma":[0.0008637294,0.00008095198,0.002362838,0.0005652094,0.0001042713,0.000006757327,0.00001539963,0.9735017,0.005677003,0.01482881,0.001213416,0.0007798762],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008299994,0.0001346371,0.9852015,0.0009920027,0.001844071,0.001389697,0.00004640968,0.001352061,0.0007395682],"genre_scores_gemma":[0.5055228,0.00008476053,0.4915281,0.0005878231,0.0001803342,0.0005792858,0.0001819975,0.00004361538,0.001291247],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6403694,"threshold_uncertainty_score":0.9997559,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08142263447052059,"score_gpt":0.3227335214040331,"score_spread":0.2413108869335125,"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."}}