{"id":"W3044911731","doi":"10.1007/978-3-030-58539-6_33","title":"A Unifying Mutual Information View of Metric Learning: Cross-Entropy vs. Pairwise Losses","year":2020,"lang":"en","type":"preprint","venue":"Lecture notes in computer science","topic":"Face recognition and analysis","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Pairwise comparison; Mutual information; Cross entropy; Entropy (arrow of time); Upper and lower bounds; Heuristics; Discriminative model; Computer science; Mathematics; Metric (unit); Mathematical optimization; Artificial intelligence; Principle of maximum entropy; Engineering","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":[0.008223846,0.00217686,0.002417454,0.002269962,0.0005507307,0.00378187,0.003188251,0.002396189,0.001763495],"category_scores_gemma":[0.02095728,0.000611932,0.001337284,0.002100447,0.002910054,0.01028181,0.004675647,0.004344799,0.0004953626],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001226575,"about_ca_system_score_gemma":0.001101429,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007032208,"about_ca_topic_score_gemma":0.0005497791,"domain_scores_codex":[0.9930325,0.003853382,0.0003467081,0.0006792761,0.001853201,0.0002349312],"domain_scores_gemma":[0.9908269,0.005442844,0.0006934949,0.001710896,0.0009984183,0.0003275097],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001828961,0.0001157259,0.001181584,0.0002696387,0.0002041572,0.0001067527,0.0001582431,0.1214086,0.002709519,0.7464732,0.003498607,0.1236911],"study_design_scores_gemma":[0.0000150236,0.0001631971,0.0003808674,0.00004109649,0.00004143161,0.0001042015,0.00002391424,0.4635147,0.001275912,0.5324706,0.001931416,0.00003746684],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008272121,0.001733063,0.9859927,0.0007222266,0.0001002804,0.00003218608,0.000106271,0.00009051126,0.002950703],"genre_scores_gemma":[0.6953052,0.005890435,0.2886085,0.0007667512,0.001917349,0.0003949824,0.0007357937,0.0003855697,0.005995445],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008223846,"threshold_uncertainty_score":0.04349238,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02329649942509552,"score_gpt":0.2830448434949603,"score_spread":0.2597483440698648,"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."}}