{"id":"W4286902830","doi":"10.48550/arxiv.2110.08232","title":"Fire Together Wire Together: A Dynamic Pruning Approach with Self-Supervised Mask Prediction","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Compute Canada","keywords":"FLOPS; Hyperparameter; Computer science; Regularization (linguistics); Artificial intelligence; Reduction (mathematics); Machine learning; Pruning; Artificial neural network; Pattern recognition (psychology); Algorithm; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008539989,0.00130054,0.001461786,0.0006806474,0.0007028161,0.0009508314,0.003535033,0.001905602,0.001992124],"category_scores_gemma":[0.002556828,0.0008474931,0.0008416272,0.0006035371,0.0009686102,0.002498469,0.001840064,0.001869272,0.0009702219],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008624015,"about_ca_system_score_gemma":0.001177233,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004383148,"about_ca_topic_score_gemma":0.00743564,"domain_scores_codex":[0.9994239,0.00009933015,0.00002257284,0.0001924614,0.0001689223,0.00009293114],"domain_scores_gemma":[0.9988694,0.0003575233,0.0001226276,0.0003475521,0.0002144026,0.00008849287],"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":[0.00036408,0.000348477,0.00247563,0.00009763092,0.000221849,0.0003389258,0.0003079616,0.4843289,0.02476279,0.01108262,0.01177374,0.4638974],"study_design_scores_gemma":[0.000008311336,0.00002667291,0.0001306227,0.000004688271,0.00001175165,0.00003398447,0.000009566745,0.9923524,0.002551787,0.004324338,0.0005406356,0.000005273397],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05009709,0.0004991561,0.9421409,0.0004588533,0.00008503597,0.00007338064,0.0001111986,0.003981552,0.002552775],"genre_scores_gemma":[0.7142951,0.0002206944,0.2749574,0.0006817963,0.0001563342,0.0001768064,0.0005410487,0.0006561994,0.008314602],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004383148,"threshold_uncertainty_score":0.008715272,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03543727667811387,"score_gpt":0.1717486076622883,"score_spread":0.1363113309841745,"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."}}