{"id":"W3101629288","doi":"","title":"Delta-STN: Efficient Bilevel Optimization for Neural Networks using Structured Response Jacobians","year":2020,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Hyperparameter; Computer science; Mathematical optimization; Bilevel optimization; Artificial neural network; Jacobian matrix and determinant; Perceptron; Artificial intelligence; Machine learning; Optimization problem; Algorithm; Mathematics; Applied mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.0001006785,0.0001792707,0.0001611236,0.00007946215,0.0003359739,0.00006121483,0.0007803597,0.00008228756,0.000008820076],"category_scores_gemma":[0.00005609026,0.0002037028,0.00009901186,0.001338305,0.00006497346,0.0003235988,0.0002506968,0.0001405993,0.000003134185],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001345022,"about_ca_system_score_gemma":0.00004894483,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004498437,"about_ca_topic_score_gemma":0.000002810987,"domain_scores_codex":[0.9986289,0.000104953,0.0001650332,0.0006912253,0.00006254747,0.000347402],"domain_scores_gemma":[0.9988334,0.000229677,0.0001460491,0.0004551949,0.0001302446,0.0002054239],"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":[0.0001947593,0.00001553119,0.00005826222,0.000003750629,0.000008929097,0.00001524784,0.00007846341,0.982288,0.0004202519,0.01630527,0.00005794125,0.0005536088],"study_design_scores_gemma":[0.000539276,0.0000763665,0.0001149956,0.000004459434,0.00002254341,0.000005085531,0.00001941558,0.9981048,0.0001623382,0.0004997705,0.0002305544,0.0002204545],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2037106,0.00001479079,0.7951384,0.0004077145,0.0001158247,0.000392213,0.00001097429,0.0002018337,0.000007681121],"genre_scores_gemma":[0.9273601,0.000005040551,0.07193038,0.0005718827,0.00008214205,0.000001778564,0.000009139787,0.00001823925,0.00002137029],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7236494,"threshold_uncertainty_score":0.8306757,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08668726734114977,"score_gpt":0.2088544038859706,"score_spread":0.1221671365448208,"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."}}