{"id":"W2952455134","doi":"10.48550/arxiv.1705.04301","title":"A Feature Embedding Strategy for High-level CNN representations from Multiple ConvNets","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"","keywords":"Convolutional neural network; Artificial intelligence; Softmax function; Computer science; Pattern recognition (psychology); Embedding; Feature (linguistics); Bottleneck; Contextual image classification; Categorization; Image (mathematics); Exploit; Feature vector; Feature extraction","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":[0.0003692125,0.0009576849,0.0004882161,0.0005111646,0.0002317958,0.0004678341,0.0009977496,0.0005264028,0.002815048],"category_scores_gemma":[0.0008510238,0.000297804,0.0005765123,0.0005021311,0.0003158877,0.001329476,0.001053597,0.0009942387,0.001070255],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004808318,"about_ca_system_score_gemma":0.0005223219,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002434942,"about_ca_topic_score_gemma":0.003968453,"domain_scores_codex":[0.9998327,0.00002224003,0.00001360172,0.00004920989,0.00005539712,0.00002677573],"domain_scores_gemma":[0.9998153,0.00002862372,0.00002188322,0.00004344637,0.00007074412,0.00001990687],"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":[0.0001707596,0.0001680019,0.00093171,0.0001323976,0.0001084096,0.0001938146,0.00008842474,0.2148714,0.05210439,0.03063728,0.007587738,0.6930056],"study_design_scores_gemma":[0.000006975815,0.00007923226,0.000287149,0.000009626458,0.00001637744,0.00008449125,0.00001261197,0.9807329,0.008831744,0.007235533,0.002694682,0.00000859078],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01600372,0.000171649,0.980683,0.00009701424,0.00004662556,0.00004417039,0.00009056902,0.0009535324,0.001909818],"genre_scores_gemma":[0.5491421,0.0002746592,0.4406257,0.0001741082,0.00007993414,0.0002112197,0.0008499338,0.0001981565,0.008444246],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002815048,"threshold_uncertainty_score":0.009417236,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2167642944721717,"score_gpt":0.3049114442689005,"score_spread":0.08814714979672886,"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."}}