{"id":"W3176678760","doi":"10.36227/techrxiv.14852652.v1","title":"Deep Clustering with Self-supervision using Pairwise Data Similarities","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Face and Expression Recognition","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Cluster analysis; Pairwise comparison; Similarity (geometry); Data mining; Computer science; Autoencoder; Embedding; Data point; Clustering high-dimensional data; Correlation clustering; Pattern recognition (psychology); Artificial intelligence; Single-linkage clustering; CURE data clustering algorithm; Deep learning; Image (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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.0002647056,0.0002826915,0.0002993139,0.0001151773,0.0001725123,0.001048443,0.001931335,0.0002251814,0.0001078857],"category_scores_gemma":[0.00001918638,0.0002271473,0.00005565948,0.0001644318,0.00002195334,0.001206683,0.01138953,0.0004275968,0.00001181924],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004988384,"about_ca_system_score_gemma":0.0002293511,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002638672,"about_ca_topic_score_gemma":0.0002295803,"domain_scores_codex":[0.9978169,0.0001097132,0.0002746431,0.001071438,0.0004384964,0.0002887981],"domain_scores_gemma":[0.9971467,0.00006048986,0.0001058006,0.002414254,0.0001665921,0.0001061373],"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":[0.0001571633,0.001949311,0.003522305,0.008279768,0.001333743,0.002484258,0.02724324,0.4568084,0.006095137,0.0007373768,0.01340463,0.4779846],"study_design_scores_gemma":[0.0001854576,0.00002015259,0.0000528999,0.0007710424,0.00002836528,0.00004174995,0.0004981419,0.9967934,0.0006375,0.0001532242,0.0004509261,0.0003671128],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02046001,0.0003765755,0.9763849,0.0003652234,0.0005846415,0.0002117852,0.000009697803,0.0004723703,0.001134772],"genre_scores_gemma":[0.06640749,0.000248456,0.9322078,0.0005927947,0.0001211608,0.00001186453,0.0003075726,0.00002562741,0.00007724597],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.539985,"threshold_uncertainty_score":0.9999886,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07504930655030552,"score_gpt":0.2818927525781412,"score_spread":0.2068434460278357,"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."}}