{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001681271,0.0005914774,0.000672061,0.000933916,0.0004950445,0.002028933,0.0007760932,0.001043867,0.002692122],"category_scores_gemma":[0.005962193,0.0004526625,0.0005748692,0.0003893859,0.001953305,0.002603885,0.001609529,0.001802166,0.0004495434],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001652025,"about_ca_system_score_gemma":0.000533776,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001179739,"about_ca_topic_score_gemma":0.0005731042,"domain_scores_codex":[0.9992836,0.000306439,0.00002674184,0.0001256079,0.000182732,0.00007497746],"domain_scores_gemma":[0.9984466,0.0008778293,0.0001781948,0.0001292956,0.0002266907,0.0001413039],"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":[0.00006108997,0.00003144,0.001240077,0.0000965939,0.00002753926,0.00009340335,0.0001363809,0.2600903,0.00227659,0.7156364,0.002960432,0.01734971],"study_design_scores_gemma":[0.000009169728,0.00003828533,0.0007242744,0.00002719828,0.000005344449,0.00005766569,0.00003256748,0.5772204,0.0005261067,0.4190794,0.00226505,0.00001460138],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1701197,0.003383697,0.7918957,0.004688934,0.00007830131,0.00006990986,0.000458163,0.0005066918,0.02879895],"genre_scores_gemma":[0.9225852,0.00201121,0.06620514,0.0004084737,0.0001453321,0.0001800771,0.0004271787,0.0002777275,0.007759646],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002692122,"threshold_uncertainty_score":0.01198637,"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."}}