{"id":"W4287644884","doi":"10.48550/arxiv.2010.03533","title":"Gradient Flow in Sparse Neural Networks and How Lottery Tickets Win","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Initialization; Computer science; Generalization; Pruning; Artificial intelligence; Inference; Artificial neural network; Flow (mathematics); Machine learning; Balanced flow; Sparse matrix; Lottery; 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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001994225,0.0002711045,0.0003021275,0.0002231794,0.00009221674,0.0002514448,0.0008500222,0.0001810619,0.00001004851],"category_scores_gemma":[0.00003722395,0.0003253041,0.0001010162,0.000518876,0.00009408372,0.0003553116,0.001461209,0.0007972725,0.00001664958],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009895523,"about_ca_system_score_gemma":0.00005526131,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003023576,"about_ca_topic_score_gemma":0.00005032629,"domain_scores_codex":[0.9982092,0.0002178349,0.0001349375,0.0009961253,0.0000877219,0.0003541357],"domain_scores_gemma":[0.9989328,0.00009809829,0.0001552673,0.0005302064,0.00004331496,0.000240269],"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":[0.00002765852,0.00003558736,0.008107555,0.00003774153,0.00002360568,0.001160393,0.0007025874,0.9673736,0.000006782533,0.01795368,0.0003377578,0.004233061],"study_design_scores_gemma":[0.0004734509,0.00003733523,0.01174228,0.00004922991,0.00001538362,0.00000739808,0.00009773951,0.9840192,0.000002347409,0.002679748,0.0005568179,0.0003191276],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2426013,0.00009811639,0.7544154,0.001378906,0.0005929151,0.0002168601,0.000002152763,0.0001692632,0.0005250667],"genre_scores_gemma":[0.9967682,0.0001130589,0.00213251,0.000503041,0.00007608137,7.100141e-7,0.00001084768,0.00001537234,0.0003801925],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7541668,"threshold_uncertainty_score":0.9999199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08419623647118654,"score_gpt":0.1781086397888582,"score_spread":0.09391240331767167,"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."}}