{"id":"W7133004520","doi":"","title":"Optimization and Loss Landscape Geometry of Deep Learning","year":2022,"lang":"","type":"dissertation","venue":"TSpace","topic":"Stochastic Gradient Optimization Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Deep learning; Artificial neural network; Deep neural networks; Set (abstract data type); Class (philosophy); Convolutional neural network; Optimization problem; Focus (optics)","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0005633869,0.0004493936,0.0006017803,0.000987641,0.0004618902,0.0001752472,0.0007628435,0.0002929157,0.002049582],"category_scores_gemma":[0.0004468905,0.0005452678,0.0001209564,0.001889171,0.0001064032,0.0003323963,0.0004309771,0.0006749215,0.000004161759],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008060726,"about_ca_system_score_gemma":0.0001242442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007363804,"about_ca_topic_score_gemma":0.000004738729,"domain_scores_codex":[0.9970057,0.0002312652,0.0006622159,0.0008912905,0.0008050259,0.0004044996],"domain_scores_gemma":[0.9974148,0.0003176754,0.001131729,0.0005570436,0.0004273344,0.0001514186],"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":[0.00006575118,0.0001280488,0.002403183,0.0003952585,0.00008789843,0.0000120502,0.03022982,0.9474655,0.00009725323,0.008734781,0.0000606015,0.01031986],"study_design_scores_gemma":[0.0004557584,0.0005160669,0.0006458588,0.0001324093,0.00009402702,0.00002897971,0.006879759,0.9896504,0.0007135714,0.0001747682,0.0001810377,0.0005274243],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01401984,0.002417543,0.9789621,0.0001058184,0.0007070327,0.0005647328,0.000003640945,0.0002468289,0.002972452],"genre_scores_gemma":[0.664961,0.0026989,0.3231758,0.00006374392,0.00009735212,0.0001739347,0.0008643922,0.0001490444,0.007815732],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6557863,"threshold_uncertainty_score":0.9996999,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009943714337835304,"score_gpt":0.2898736506689021,"score_spread":0.2799299363310668,"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."}}