{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001354989,0.001288212,0.001330837,0.001369196,0.0007289078,0.001300124,0.002480844,0.001458107,0.001660521],"category_scores_gemma":[0.003827228,0.0008675416,0.001214684,0.00141509,0.001565364,0.003310509,0.002876017,0.002062503,0.0008680608],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00146247,"about_ca_system_score_gemma":0.001443355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004929717,"about_ca_topic_score_gemma":0.008233322,"domain_scores_codex":[0.9988279,0.0002390644,0.00006817899,0.0004550924,0.0003044003,0.000105386],"domain_scores_gemma":[0.9982476,0.0003855183,0.0003061166,0.0005674311,0.0003831611,0.0001101687],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002089867,0.0001992811,0.003151549,0.0001928072,0.0002175832,0.00009508736,0.0003138493,0.6333145,0.01866302,0.02162965,0.00326757,0.3187461],"study_design_scores_gemma":[0.000005749973,0.00003146025,0.0002889184,0.000007739148,0.000009089129,0.00002905485,0.00002037141,0.9864846,0.003515551,0.009157144,0.0004412172,0.00000913205],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01673621,0.0001069136,0.9817091,0.00007801632,0.00001239475,0.00003100707,0.00006005412,0.000662834,0.0006035299],"genre_scores_gemma":[0.5479146,0.0002454777,0.4468167,0.0001635515,0.00005324937,0.0001502318,0.0008519204,0.0002767956,0.003527432],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004929717,"threshold_uncertainty_score":0.01061106,"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."}}