{"id":"W4380885324","doi":"10.36227/techrxiv.23518110","title":"Communication Efficient Privacy-Preserving Federated Learning via Knowledge Distillation for Human Activity Recognition Systems","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University; Lakehead University","funders":"","keywords":"Computer science; Machine learning; Federated learning; Artificial intelligence; Distillation; Internet of Things; Mean squared error; Activity recognition; Embedded system; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002556462,0.0007380233,0.001457312,0.0005413022,0.0008116694,0.001674682,0.002128201,0.001406535,0.001138735],"category_scores_gemma":[0.007406035,0.0003862007,0.0007305542,0.001066554,0.001351926,0.003612249,0.003173444,0.001907948,0.0003128547],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001222123,"about_ca_system_score_gemma":0.0017855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002855498,"about_ca_topic_score_gemma":0.002136953,"domain_scores_codex":[0.9981177,0.0005968453,0.0001327681,0.0004739885,0.0004180196,0.0002606834],"domain_scores_gemma":[0.9964705,0.001749965,0.0003313879,0.0009330966,0.0003623431,0.000152796],"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":[0.0005001156,0.0003184146,0.001189375,0.00009453583,0.00007685284,0.0001566314,0.0001343855,0.7578196,0.004317143,0.01394944,0.002333772,0.2191098],"study_design_scores_gemma":[0.00001098963,0.0000359637,0.00009661767,0.000002919062,0.000004890743,0.00002927469,0.00001268453,0.9886742,0.00180546,0.009104358,0.0002167726,0.000005908765],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04632026,0.0003850781,0.9507015,0.0003935714,0.00003989764,0.0000444572,0.0001314855,0.001016766,0.0009669948],"genre_scores_gemma":[0.9133457,0.0001322312,0.08443497,0.0002514671,0.00002820683,0.00007431128,0.000310636,0.00005067468,0.001371807],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002855498,"threshold_uncertainty_score":0.01352,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1137490068846616,"score_gpt":0.3416331261215287,"score_spread":0.2278841192368671,"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."}}