{"id":"W4414387560","doi":"10.3390/ai6090238","title":"GRU-BERT for NILM: A Hybrid Deep Learning Architecture for Load Disaggregation","year":2025,"lang":"en","type":"article","venue":"AI","topic":"Smart Grid Energy Management","field":"Engineering","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Deep learning; Energy consumption; Software deployment; Transformer; Key (lock); Cloud computing; Encoder; Autoencoder; Architecture","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":[],"consensus_categories":[],"category_scores_codex":[0.00009166977,0.0001054479,0.0001021922,0.00006999364,0.00007106242,0.00003151447,0.00009382543,0.00002978915,0.000009610817],"category_scores_gemma":[0.00008546102,0.0001096271,0.00006865053,0.00009226594,0.000009077698,0.0000496906,0.00002567207,0.00008397608,0.000007454977],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009200932,"about_ca_system_score_gemma":0.000009017699,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001024977,"about_ca_topic_score_gemma":0.00006357321,"domain_scores_codex":[0.9994632,0.000006836151,0.000115454,0.0001443585,0.00007137759,0.0001987623],"domain_scores_gemma":[0.999699,0.0000923154,0.00001446311,0.0001388203,0.00003206109,0.00002332187],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00001981699,0.000008291607,0.0001962189,0.0002064964,0.00009155965,7.003692e-7,0.0001062316,0.8707357,0.0003396164,0.00342374,0.02192981,0.1029418],"study_design_scores_gemma":[0.0005437054,0.00003431123,0.0004990694,0.00004883981,0.0000412019,7.533957e-7,0.00001774168,0.3673179,0.004010454,0.004427294,0.6229136,0.0001451072],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01155213,0.0003015833,0.9805287,0.0009656107,0.0008248491,0.0004113453,0.000004019337,0.0003442344,0.005067545],"genre_scores_gemma":[0.9882289,0.00002458097,0.00600605,0.0005055511,0.0002931475,0.0005547968,0.00007942104,0.00004748406,0.004260061],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9766768,"threshold_uncertainty_score":0.4470463,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.003842496880986651,"score_gpt":0.2133682348081186,"score_spread":0.2095257379271319,"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."}}