{"id":"W2976931991","doi":"10.48550/arxiv.1805.11123","title":"Global Sum Pooling: A Generalization Trick for Object Counting with\\n Small Datasets of Large Images","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Video Surveillance and Tracking Methods","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"","keywords":"Overfitting; Pooling; Computer science; Generalization; Artificial intelligence; Convolutional neural network; Inference; Pattern recognition (psychology); Focus (optics); Object (grammar); Image (mathematics); Property (philosophy); Machine learning; Artificial neural network; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004895484,0.004035068,0.002411556,0.001435705,0.0007749607,0.001908786,0.005831012,0.002261781,0.003776581],"category_scores_gemma":[0.008088093,0.001261838,0.00266653,0.002471496,0.001710916,0.005604783,0.003629066,0.004223003,0.002228882],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001519875,"about_ca_system_score_gemma":0.001545772,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009102057,"about_ca_topic_score_gemma":0.01152968,"domain_scores_codex":[0.9983662,0.000272805,0.00009906986,0.0007876807,0.0003008124,0.0001734512],"domain_scores_gemma":[0.9978613,0.0006580089,0.0002004348,0.0009864261,0.0002035644,0.00009026039],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003949777,0.000221531,0.003613485,0.0004176607,0.0007483282,0.0003730599,0.0002224685,0.2167838,0.02114165,0.01606802,0.01479532,0.7252196],"study_design_scores_gemma":[0.00002293608,0.0001127027,0.001273678,0.00003905281,0.000100304,0.0001585756,0.00004305148,0.960797,0.0109979,0.02186799,0.004556577,0.00003024735],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01337109,0.0005657122,0.9787201,0.0003721443,0.00009750622,0.0001210229,0.0005313216,0.004961011,0.001260205],"genre_scores_gemma":[0.3119873,0.001149824,0.6703837,0.001273376,0.000414362,0.0006176917,0.004109428,0.001022429,0.009041814],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009102057,"threshold_uncertainty_score":0.02589005,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.085670418535971,"score_gpt":0.248886750074821,"score_spread":0.16321633153885,"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."}}