{"id":"W4412006979","doi":"10.2139/ssrn.5337603","title":"Ayla: Amplifying Gradient Sensitivity Via Loss Transformation in Non-Convex Optimization","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Sparse and Compressive Sensing Techniques","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Sensitivity (control systems); Transformation (genetics); Regular polygon; Convex optimization; Mathematics; Mathematical optimization; Computer science; Geometry; Chemistry; Electronic engineering; Engineering","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","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001013467,0.0003339123,0.0004127025,0.0004762636,0.00009879321,0.00009102283,0.0001896825,0.0003328062,0.000003909278],"category_scores_gemma":[0.00001370402,0.0003686906,0.0001667744,0.0002152382,0.00002394126,0.0001850383,0.00006929558,0.004015964,0.000002649327],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001883905,"about_ca_system_score_gemma":0.0004821326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001572147,"about_ca_topic_score_gemma":0.0006885077,"domain_scores_codex":[0.9976037,0.0001033048,0.0004947206,0.0002412405,0.0002145811,0.001342486],"domain_scores_gemma":[0.9994302,0.00003780012,0.0001277351,0.0002506858,0.0001021249,0.00005144066],"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":[0.00001907247,0.00002463155,0.00007663312,0.00007232654,0.0001266276,0.00001601781,0.00038943,0.975803,0.0003680021,0.001272724,0.000026701,0.02180489],"study_design_scores_gemma":[0.0003677832,0.00003764318,0.0001495594,0.0005968382,0.00005936728,0.0002563349,0.0001487234,0.9471101,0.003008632,0.04778952,0.00006410221,0.000411369],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06347033,0.0009441543,0.9331404,0.0001214188,0.000582367,0.000337082,0.000006003361,0.0002888544,0.001109379],"genre_scores_gemma":[0.9903613,0.007490604,0.001834051,0.00003111325,0.0001601133,0.00001575182,0.00004116825,0.00003623475,0.00002968088],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9313064,"threshold_uncertainty_score":0.9998765,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007461821204500602,"score_gpt":0.2276346138058728,"score_spread":0.2201727926013722,"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."}}