{"id":"W4385834296","doi":"10.1109/access.2023.3305385","title":"RandMixAugment: A Novel Unified Technique for Region- and Image-Level Data Augmentations","year":2023,"lang":"en","type":"article","venue":"IEEE Access","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"Korea Agency for Infrastructure Technology Advancement; Ministry of Land, Infrastructure and Transport","keywords":"Computer science; Generalization; Artificial intelligence; Deep learning; Image (mathematics); Masking (illustration); Machine learning; Data modeling; Contextual image classification; Pattern recognition (psychology); Mathematics; Database","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":[0.001134908,0.001780205,0.001250078,0.001002405,0.0004911361,0.0009814024,0.002551119,0.00120429,0.002644416],"category_scores_gemma":[0.003270107,0.0006645819,0.001497382,0.001082132,0.001423349,0.002818931,0.004599433,0.002416134,0.001216522],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005093951,"about_ca_system_score_gemma":0.001064232,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001746626,"about_ca_topic_score_gemma":0.002521037,"domain_scores_codex":[0.9991535,0.0001716515,0.00004746795,0.0002601333,0.0002777965,0.00008942219],"domain_scores_gemma":[0.9991664,0.0001789118,0.00009612169,0.0004184001,0.0000894851,0.0000507455],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004939163,0.0002277858,0.001575977,0.0002469189,0.000165321,0.0002973132,0.0003922234,0.1685143,0.0892103,0.03084236,0.01184033,0.6961933],"study_design_scores_gemma":[0.00003093416,0.0001381585,0.0003340864,0.00002632315,0.00003539176,0.0002965561,0.00003570327,0.9220383,0.0534169,0.01250379,0.01110367,0.00004030549],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01017751,0.0003279117,0.9848909,0.0001272906,0.00005857207,0.0000605444,0.0001111216,0.003255419,0.000990825],"genre_scores_gemma":[0.2063176,0.0004567065,0.7870104,0.0003161276,0.0001023653,0.000291677,0.0007067975,0.0006593274,0.00413892],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002644416,"threshold_uncertainty_score":0.008846462,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2366283297808449,"score_gpt":0.4054351063097579,"score_spread":0.1688067765289129,"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."}}