{"id":"W4283204791","doi":"10.1109/infocom48880.2022.9796896","title":"Distributed Inference with Deep Learning Models across Heterogeneous Edge Devices","year":2022,"lang":"en","type":"article","venue":"IEEE INFOCOM 2022 - IEEE Conference on Computer Communications","topic":"IoT and Edge/Fog Computing","field":"Computer Science","cited_by":70,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Inference; Computer science; Deep learning; Directed acyclic graph; Artificial intelligence; Approximate inference; Enhanced Data Rates for GSM Evolution; Edge device; Machine learning; Theoretical computer science; Algorithm","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":[0.001128591,0.001304682,0.0008298383,0.0004885438,0.0006681803,0.001281931,0.002726843,0.0009517397,0.002903072],"category_scores_gemma":[0.003045581,0.0006883439,0.0007577455,0.0005761212,0.0007455546,0.003569578,0.001800965,0.002558935,0.0005723801],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001566988,"about_ca_system_score_gemma":0.002054841,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01407338,"about_ca_topic_score_gemma":0.02171577,"domain_scores_codex":[0.9993647,0.0001044935,0.00003372165,0.0002523621,0.0001298464,0.0001148882],"domain_scores_gemma":[0.9991092,0.0003654622,0.00004922712,0.0002597003,0.0001518054,0.00006459714],"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":[0.000644659,0.0003313449,0.003824452,0.0001146213,0.0001577905,0.0002484595,0.0001467468,0.7867822,0.008046677,0.01144196,0.007072239,0.1811888],"study_design_scores_gemma":[0.00001331829,0.00002142083,0.00009481512,0.000003351561,0.000008830366,0.00001032749,0.00001224701,0.9932127,0.002104873,0.004093723,0.0004196792,0.000004643656],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0632492,0.0004251754,0.9262747,0.000527334,0.0001188051,0.00008189137,0.0002790933,0.006500021,0.00254374],"genre_scores_gemma":[0.7896983,0.0002676819,0.2051944,0.0004186039,0.00004460097,0.000104235,0.0006887308,0.0002491177,0.003334357],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01407338,"threshold_uncertainty_score":0.02798289,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08018509552167943,"score_gpt":0.3042089167860083,"score_spread":0.2240238212643289,"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."}}