{"id":"W3194068170","doi":"10.1007/s41884-021-00053-7","title":"Pseudo-Riemannian geometry encodes information geometry in optimal transport","year":2021,"lang":"en","type":"article","venue":"Information Geometry","topic":"Geometric Analysis and Curvature Flows","field":"Mathematics","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; University of Southern California; Connaught Fund; National Science Foundation","keywords":"Information geometry; Mathematics; Geodesic; Statistical manifold; Geometry; Riemannian geometry; Differential geometry; Constant curvature; Geometric flow; Curvature; Parallel transport; Divergence (linguistics); Convex geometry; Scalar curvature; Regular polygon","routes":{"ca_aff":true,"ca_fund":true,"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","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001737125,0.0005242684,0.0008667063,0.004443603,0.0002350878,0.0003138431,0.0005472074,0.0005563861,0.002058296],"category_scores_gemma":[0.001931761,0.000519884,0.0004352817,0.0111458,0.00008405632,0.008162762,0.0001495292,0.0008239105,0.0008510067],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002736504,"about_ca_system_score_gemma":0.0002924937,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007115927,"about_ca_topic_score_gemma":0.00005012694,"domain_scores_codex":[0.9951424,0.00009424114,0.002117052,0.0003252784,0.001470284,0.0008507635],"domain_scores_gemma":[0.9968656,0.0003920805,0.0007987089,0.0008890858,0.0007628812,0.0002916669],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0006410738,0.002996986,0.4368439,0.006911735,0.002317216,0.0003536894,0.03028088,0.03464978,0.0004073543,0.136656,0.06833769,0.2796038],"study_design_scores_gemma":[0.01071978,0.0004947403,0.4788444,0.0008045979,0.0008478037,0.0006250017,0.04361499,0.02555452,0.005772065,0.01149554,0.4160798,0.00514672],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9186361,0.0002682506,0.0634358,0.000440893,0.0005554499,0.0004320663,0.0001497179,0.0002605195,0.01582123],"genre_scores_gemma":[0.9786388,0.0001574998,0.01813194,0.001126132,0.000147093,0.00006581883,0.001116672,0.00003102908,0.0005849849],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3477421,"threshold_uncertainty_score":0.9999269,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01425632435491615,"score_gpt":0.2567735146344364,"score_spread":0.2425171902795203,"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."}}