{"id":"W4385844880","doi":"10.2139/ssrn.4538377","title":"A Transfer Learning Approach to Minimize Reinforcement Learning Risks in Energy Optimization for Smart Buildings","year":2023,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Building Energy and Comfort Optimization","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Reinforcement learning; Transfer of learning; Reinforcement; Computer science; Energy (signal processing); Architectural engineering; Energy transfer; Artificial intelligence; Engineering; Structural engineering; Engineering physics; 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":[0.00229713,0.00100021,0.001589585,0.0004444679,0.0003747831,0.0008481697,0.00150837,0.002314135,0.004481749],"category_scores_gemma":[0.007462307,0.0006075597,0.0006081536,0.0004956309,0.001385427,0.00168708,0.00187916,0.002246554,0.0003550769],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001009709,"about_ca_system_score_gemma":0.001113435,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004238781,"about_ca_topic_score_gemma":0.00230056,"domain_scores_codex":[0.9993716,0.0003440557,0.00002318923,0.00008707176,0.0001126123,0.00006151783],"domain_scores_gemma":[0.9973531,0.002103115,0.0001193475,0.00008352999,0.0002498954,0.0000910056],"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":[0.00003309565,0.00003869522,0.0001346891,0.00003849991,0.00002283257,0.00002218,0.00002571352,0.9774746,0.0002517961,0.009229722,0.0004565209,0.01227166],"study_design_scores_gemma":[0.000005522076,0.00001287726,0.00002035499,0.000003050095,0.000002497479,0.000002130333,0.000002126201,0.9947122,0.00004709872,0.005115219,0.00007515931,0.000001859327],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01715526,0.0003110402,0.9792442,0.0003840134,0.00004991335,0.00004736272,0.00002625074,0.0001295018,0.002652477],"genre_scores_gemma":[0.8752241,0.0003927449,0.1128442,0.0002207383,0.0001728327,0.0003247737,0.00008763046,0.0001575681,0.01057537],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004481749,"threshold_uncertainty_score":0.01499301,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02003663576072724,"score_gpt":0.2346075748484468,"score_spread":0.2145709390877196,"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."}}