{"id":"W4313436112","doi":"10.3390/s23010006","title":"Federated Learning via Augmented Knowledge Distillation for Heterogenous Deep Human Activity Recognition Systems","year":2022,"lang":"en","type":"article","venue":"Sensors","topic":"Context-Aware Activity Recognition Systems","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thunder Bay Regional Research Institute; Lakehead University","funders":"","keywords":"Computer science; Distillation; Activity recognition; Deep learning; Artificial intelligence; Human–computer interaction; Machine learning; Embedded system; Chemistry; Chromatography","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","sts"],"consensus_categories":[],"category_scores_codex":[0.0006839004,0.0002253649,0.0003146451,0.0002321433,0.001701515,0.0003136646,0.0002494733,0.0000730801,0.00002933304],"category_scores_gemma":[0.00007491236,0.0002651779,0.0001210156,0.0004919667,0.00002492192,0.0003699578,0.0002031301,0.0002996767,0.00007556283],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004513902,"about_ca_system_score_gemma":0.00004896609,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002656035,"about_ca_topic_score_gemma":0.0001645369,"domain_scores_codex":[0.9973926,0.0009233798,0.0003553339,0.0006122322,0.0003344541,0.0003820276],"domain_scores_gemma":[0.9986646,0.0003333746,0.000354146,0.0002511596,0.0002919027,0.0001048019],"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":[0.0002195051,0.0009440525,0.001131207,0.0003382906,0.0003685608,0.00004310213,0.005421994,0.0140531,0.1143175,0.0001997726,0.0004095707,0.8625534],"study_design_scores_gemma":[0.001194028,0.000627482,0.0008647506,0.00003931707,0.00002920094,0.0001836894,0.0005087481,0.9834252,0.004688763,0.0001582308,0.007736053,0.000544525],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6540385,0.0000605942,0.3429018,0.00005614643,0.0009037247,0.0009151975,0.00002625874,0.0004781,0.0006196442],"genre_scores_gemma":[0.9984893,0.00000117001,0.0001544968,0.00001659823,0.0001416223,0.0003900884,0.0001395582,0.00003495709,0.0006321766],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9693721,"threshold_uncertainty_score":0.99998,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04525981913289185,"score_gpt":0.2785962715836858,"score_spread":0.2333364524507939,"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."}}