{"id":"W4405268798","doi":"10.36227/techrxiv.173396113.31607552/v1","title":"Impact of Inaccurate Contamination Ratios on Robust Unsupervised Anomaly Detection: Experimental Investigation","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke","funders":"","keywords":"Contamination; Anomaly detection; Anomaly (physics); Computer science; Benchmark (surveying); Misinformation; Data mining; Artificial intelligence; Geography; Biology; Ecology; Cartography","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002064467,0.0003181467,0.0002873874,0.0004196013,0.0001159405,0.0002686607,0.0005913876,0.0002833939,0.00005877796],"category_scores_gemma":[0.00001602604,0.0002825971,0.0002953507,0.0005638769,0.00006842549,0.0002421922,0.0006481091,0.0004362309,0.0000543612],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004325406,"about_ca_system_score_gemma":0.000243708,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003735687,"about_ca_topic_score_gemma":0.00002575041,"domain_scores_codex":[0.998197,0.00008453167,0.0005363359,0.0007078425,0.0003024951,0.0001717576],"domain_scores_gemma":[0.9985147,0.00004970321,0.0003041837,0.0008118651,0.0002222079,0.00009735218],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000124942,0.00112911,0.00175988,0.0005762062,0.0006476608,0.00001948957,0.004218124,0.02295175,0.6351944,0.2559547,0.004258659,0.07316518],"study_design_scores_gemma":[0.0002474295,0.0008629448,0.01198685,0.0001315473,0.00002564632,0.00001271182,0.00006459893,0.2229716,0.7538533,0.009345493,0.00005035139,0.0004475045],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5887244,0.00007559593,0.4073624,0.0002154066,0.0002621779,0.0007192399,0.000015033,0.0006399816,0.001985777],"genre_scores_gemma":[0.989732,0.00001367326,0.009279066,0.00006277942,0.00009675289,0.0004507669,0.00003189415,0.00002367652,0.000309342],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4010077,"threshold_uncertainty_score":0.9999626,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03669432959578724,"score_gpt":0.3019553567641288,"score_spread":0.2652610271683415,"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."}}