{"id":"W3184153358","doi":"10.1109/tpami.2022.3201541","title":"Bayesian Embeddings for Few-Shot Open World Recognition","year":2022,"lang":"en","type":"article","venue":"IEEE Transactions on Pattern Analysis and Machine Intelligence","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Stanford University; National Aeronautics and Space Administration","keywords":"Artificial intelligence; Computer science; Open set; Embedding; Machine learning; Benchmark (surveying); Shot (pellet); Bayesian probability; Measure (data warehouse); Class (philosophy); Prior probability; Parametric statistics; Data mining; Mathematics","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0006151598,0.0001909028,0.0002892021,0.000760262,0.0008663414,0.0004216735,0.0008380169,0.00002652745,0.001245801],"category_scores_gemma":[0.000006405119,0.0001957271,0.000243284,0.001710506,0.0000338376,0.000389526,0.000027125,0.0003226237,0.00002029341],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006498436,"about_ca_system_score_gemma":0.00002900995,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006151959,"about_ca_topic_score_gemma":0.001318799,"domain_scores_codex":[0.998291,0.0001554159,0.0003820698,0.0006224548,0.0002913452,0.0002577208],"domain_scores_gemma":[0.9990318,0.0002289832,0.0001583033,0.0003766525,0.0000711613,0.0001330772],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000369662,0.0001672402,0.0001138909,0.000009316941,0.0002725899,0.000004866451,0.0007377826,0.04666176,0.0001229159,0.0002698418,0.00006771239,0.9515351],"study_design_scores_gemma":[0.0003365517,0.0003183348,0.0002228372,0.00001570529,0.0003616269,0.00001681771,0.0003985111,0.9748709,0.01459805,0.002265528,0.006114542,0.0004805483],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0003682819,0.0000325707,0.9971505,0.001016376,0.000259547,0.0003408941,0.00008527195,0.00008518562,0.0006613192],"genre_scores_gemma":[0.9859475,0.00003951887,0.009891576,0.001822404,0.00001604605,0.0002642941,0.0000317445,0.00001555712,0.001971398],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.987259,"threshold_uncertainty_score":0.9996672,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05490974341790233,"score_gpt":0.3117677968617627,"score_spread":0.2568580534438604,"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."}}