{"id":"W2740792888","doi":"10.24963/ijcai.2017/463","title":"Learning Mahalanobis Distance Metric: Considering Instance Disturbance Helps","year":2017,"lang":"en","type":"article","venue":"","topic":"Face and Expression Recognition","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"Novelis (Canada)","funders":"National Natural Science Foundation of China","keywords":"Mahalanobis distance; Metric (unit); Artificial intelligence; Computer science; Feature (linguistics); Visualization; Pattern recognition (psychology); Similarity (geometry); Machine learning; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001693515,0.0001298051,0.0001546919,0.00006019495,0.000822652,0.0006659675,0.0009187738,0.00004838304,0.00004508021],"category_scores_gemma":[0.0002917993,0.0001110809,0.00005164152,0.0001280154,0.00006911995,0.001326376,0.0003468628,0.0001818724,0.0001925576],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002843151,"about_ca_system_score_gemma":0.00002485171,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004860508,"about_ca_topic_score_gemma":0.0000351325,"domain_scores_codex":[0.9988843,0.00003276187,0.0001687644,0.0003970513,0.0002385809,0.0002786054],"domain_scores_gemma":[0.9988079,0.00008112288,0.0001807538,0.0007677724,0.00007199847,0.00009045836],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004530461,0.0002028786,0.1412934,0.0001629568,0.00007251035,0.0002799917,0.001931592,0.0006218189,0.01378923,0.1225026,0.01964782,0.69945],"study_design_scores_gemma":[0.003041605,0.000301108,0.1309475,0.0009187299,0.00002271309,0.0000906127,0.0005726781,0.1001961,0.1660059,0.02145166,0.5740764,0.002374989],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06289965,0.0003350649,0.8517461,0.001758624,0.000694318,0.0001349635,0.000002138929,0.000390523,0.08203861],"genre_scores_gemma":[0.9685087,0.00009464195,0.02541164,0.0002130223,0.00004260858,0.00001504599,0.000001416588,0.000007641835,0.005705252],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9056091,"threshold_uncertainty_score":0.642194,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02185502056054667,"score_gpt":0.2575147026790977,"score_spread":0.235659682118551,"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."}}