{"id":"W3169317873","doi":"10.48550/arxiv.2106.03632","title":"Quantifying and Improving Transferability in Domain Generalization","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Transferability; Computer science; Classifier (UML); Generalization; Benchmark (surveying); Invariant (physics); Artificial intelligence; Machine learning; Domain (mathematical analysis); Algorithm; Transfer of learning; Theoretical computer science; 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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005044089,0.00019947,0.0002620346,0.000229655,0.0001280849,0.000266134,0.0004964069,0.0001958049,0.00001588101],"category_scores_gemma":[0.00004470759,0.0002546413,0.00008637116,0.0005547459,0.00005988152,0.000491122,0.0005881151,0.0004590683,0.00000289926],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001498582,"about_ca_system_score_gemma":0.0001598059,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003107933,"about_ca_topic_score_gemma":0.0005349128,"domain_scores_codex":[0.9981545,0.0003258554,0.0002165998,0.0009800887,0.00007838427,0.0002446008],"domain_scores_gemma":[0.9990957,0.00007783894,0.0001071277,0.0005359834,0.00008185719,0.0001015305],"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":[0.00003509139,0.0001592708,0.06909019,0.0004414105,0.0000521322,0.0005945925,0.007720626,0.5478283,0.002252562,0.3595307,0.000008220925,0.01228683],"study_design_scores_gemma":[0.000583547,0.00001747523,0.0203974,0.00008785892,0.00001264253,0.000004506542,0.0006320525,0.9704406,0.0000947744,0.007265161,0.0001004002,0.0003635444],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4796081,0.00006174312,0.519707,0.00005495986,0.0001375514,0.0001031591,8.373679e-7,0.0000638754,0.0002627712],"genre_scores_gemma":[0.9880872,0.0001357185,0.01155166,0.00008412275,0.0000193401,7.984825e-7,0.00001417829,0.0000106872,0.00009634077],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.508479,"threshold_uncertainty_score":0.9999906,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09783610275853746,"score_gpt":0.2052525491074624,"score_spread":0.1074164463489249,"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."}}