{"id":"W4401023759","doi":"10.24963/ijcai.2024/572","title":"Optimal Transport on Categorical Data for Conterfactuals Using Compositional Data and Dirichlet Transport","year":2024,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Université du Québec à Montréal","funders":"Office of Advanced Cyberinfrastructure; National Centre for Supercomputing Applications; National Science Foundation","keywords":"Computer science; Speech recognition; Microphone; Discriminative model; Speaker diarisation; Speaker recognition; Leverage (statistics); Context (archaeology); Artificial intelligence; Task (project management); Unsupervised learning; Reduction (mathematics); Speech enhancement; Noise reduction; Pattern recognition (psychology); 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":[0.03325782,0.001746671,0.003140862,0.003594019,0.002528081,0.005296298,0.004448303,0.004023233,0.006636025],"category_scores_gemma":[0.08362823,0.001686491,0.004849635,0.00320269,0.006841228,0.0104088,0.006130833,0.006224365,0.0009561022],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005389072,"about_ca_system_score_gemma":0.004273488,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008187527,"about_ca_topic_score_gemma":0.006805153,"domain_scores_codex":[0.9891887,0.006993937,0.0005625038,0.001921501,0.000854113,0.0004792956],"domain_scores_gemma":[0.9529877,0.03788894,0.002239599,0.004433153,0.00171361,0.0007368705],"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":[0.0002032699,0.00006742014,0.002619293,0.0002018253,0.0001077079,0.000201801,0.0006655513,0.3076517,0.000812424,0.6543666,0.0009811219,0.03212135],"study_design_scores_gemma":[0.00002013357,0.00002549544,0.0002597882,0.00004742831,0.00001896916,0.00004223642,0.00008202461,0.5628107,0.000523485,0.435046,0.001092045,0.00003170535],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008386945,0.0001504077,0.9901568,0.0003193029,0.00003126304,0.00007333,0.0001019279,0.0001172957,0.0006627658],"genre_scores_gemma":[0.3454354,0.0007481448,0.6459002,0.0004657506,0.0002048139,0.0008139162,0.001014942,0.0005710754,0.004845729],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03325782,"threshold_uncertainty_score":0.1758863,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1231895037104952,"score_gpt":0.3489885142031081,"score_spread":0.2257990104926129,"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."}}