{"id":"W4399282665","doi":"10.1109/access.2024.3409057","title":"DNNSplit: Latency and Cost-Efficient Split Point Identification for Multi-Tier DNN Partitioning","year":2024,"lang":"en","type":"article","venue":"IEEE Access","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Latency (audio); Scalability; Inference; Computational complexity theory; Deep neural networks; Artificial neural network; Parallel computing; Distributed computing; Theoretical computer science; Artificial intelligence; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002039514,0.0001184567,0.0001006115,0.0000883916,0.0002380908,0.0007208724,0.000545506,0.00004256405,0.000005870449],"category_scores_gemma":[0.00002578878,0.0001131458,0.00004535599,0.0004568816,0.00004716747,0.0008738381,0.0001346251,0.00009785959,0.00004595141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006376713,"about_ca_system_score_gemma":0.00002954334,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007519,"about_ca_topic_score_gemma":0.00000770006,"domain_scores_codex":[0.9987794,0.00001936788,0.0002658452,0.0005529507,0.0001349901,0.0002474374],"domain_scores_gemma":[0.9991995,0.0001381564,0.0000699001,0.0004180251,0.00009262718,0.00008183705],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002040485,0.0003078056,0.0008220425,0.0003704729,0.0000690559,0.00002195395,0.001627781,0.06899215,0.01812939,0.4516301,0.008387016,0.4496219],"study_design_scores_gemma":[0.0001656216,0.00001286392,0.002019546,0.00004960511,0.00001245192,0.000009628237,0.000006972169,0.9701521,0.008000794,0.01205299,0.007348238,0.0001691788],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0319345,0.000484935,0.9643573,0.001244511,0.0007039285,0.0008686214,0.0000136543,0.0003108064,0.00008174601],"genre_scores_gemma":[0.9697446,0.00005970192,0.02834781,0.0001953825,0.0001502526,0.0009664589,0.00001268408,0.0000189785,0.0005040877],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9378102,"threshold_uncertainty_score":0.6951389,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08664129922105507,"score_gpt":0.3798787161081608,"score_spread":0.2932374168871057,"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."}}