{"id":"W2895255442","doi":"10.1109/wacv.2019.00040","title":"Multi-Scale Convolution Aggregation and Stochastic Feature Reuse for DenseNets","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Memorial University of Newfoundland","funders":"","keywords":"Overfitting; Computer science; Reuse; Convolution (computer science); Feature (linguistics); Regularization (linguistics); Concatenation (mathematics); Artificial intelligence; Scale (ratio); Aggregate (composite); Convolutional neural network; Artificial neural network; Machine learning; Algorithm; Pattern recognition (psychology); 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":[0.0001210297,0.000217418,0.000215054,0.00007366567,0.0001187498,0.0001173108,0.0007912638,0.0002554216,0.000001154848],"category_scores_gemma":[0.00008630461,0.0002089773,0.00006277525,0.0001165133,0.00004322355,0.000181616,0.001274769,0.0002912877,0.00001730548],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007184003,"about_ca_system_score_gemma":0.00006280182,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007654839,"about_ca_topic_score_gemma":0.00003822902,"domain_scores_codex":[0.9986339,0.00003393763,0.0001819485,0.0007862792,0.0001369509,0.0002269544],"domain_scores_gemma":[0.9979532,0.0002537933,0.0002060659,0.001339626,0.0001659298,0.00008142624],"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":[0.00008677725,0.0002976849,0.0005645161,0.0007273681,0.0001333479,0.000003217603,0.001690111,0.7283611,0.01155284,0.04125611,0.0437981,0.1715288],"study_design_scores_gemma":[0.00035847,0.00002475616,0.001068809,0.00006697629,0.00001575498,0.000009977339,0.00000508885,0.9818388,0.0004635204,0.01490805,0.0009823054,0.0002575451],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003068157,0.0004825826,0.9914637,0.002444216,0.0004866362,0.001741234,0.00002056826,0.0002591231,0.00003373605],"genre_scores_gemma":[0.1681972,0.00005967522,0.8287579,0.0002541374,0.0001186377,0.0003601794,0.00007199828,0.00002065885,0.002159562],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2534777,"threshold_uncertainty_score":0.8521847,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02868478204061275,"score_gpt":0.2889173167260632,"score_spread":0.2602325346854504,"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."}}