{"id":"W4385019662","doi":"10.48550/arxiv.2307.10507","title":"FedSoup: Improving Generalization and Personalization in Federated Learning via Selective Model Interpolation","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"AI in cancer detection","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China; Public Safety Canada; Compute Canada","keywords":"Overfitting; Computer science; Generalization; Personalization; Code (set theory); Interpolation (computer graphics); Machine learning; Artificial intelligence; Maxima and minima; Distribution (mathematics); Data mining; Distributed computing; Image (mathematics); Artificial neural network; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004003907,0.001883363,0.002051748,0.001252034,0.0009672329,0.001339886,0.003471573,0.002208699,0.002311798],"category_scores_gemma":[0.01140539,0.0009244318,0.001545956,0.001028879,0.001274202,0.003284509,0.003740744,0.002995464,0.001060728],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00137235,"about_ca_system_score_gemma":0.0021459,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01096673,"about_ca_topic_score_gemma":0.01266339,"domain_scores_codex":[0.998409,0.0004838792,0.00009466781,0.0004914073,0.0003095421,0.0002115265],"domain_scores_gemma":[0.9959077,0.001378527,0.0002313651,0.001780026,0.0004988792,0.0002035099],"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":[0.0003526035,0.0003787388,0.004229451,0.000064894,0.000153586,0.0001931555,0.0002244792,0.7710046,0.00423352,0.003111148,0.005255556,0.2107982],"study_design_scores_gemma":[0.0000150077,0.00002411071,0.0001022379,0.00000422978,0.000007228924,0.00002102655,0.00001238205,0.9960113,0.0009901874,0.002531726,0.0002739957,0.000006559688],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04815919,0.0003233824,0.9374219,0.0003242776,0.00005989311,0.00009616969,0.000202031,0.0122878,0.001125286],"genre_scores_gemma":[0.6527795,0.0001530136,0.3411914,0.0006165162,0.00006972871,0.0002488001,0.001128697,0.0009324502,0.002879922],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01096673,"threshold_uncertainty_score":0.02180576,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05828039505646204,"score_gpt":0.1936485410933323,"score_spread":0.1353681460368703,"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."}}