{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001164232,0.001769681,0.0008614764,0.0007941774,0.0007526299,0.001335471,0.002442701,0.001170506,0.004693533],"category_scores_gemma":[0.004009461,0.0006598139,0.0007056927,0.0005353387,0.0005314418,0.002365058,0.002119813,0.001588167,0.001157227],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002190914,"about_ca_system_score_gemma":0.002489009,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01300392,"about_ca_topic_score_gemma":0.02545655,"domain_scores_codex":[0.9994627,0.0001030323,0.00004301288,0.0001376128,0.0001604261,0.0000933275],"domain_scores_gemma":[0.9990043,0.000415327,0.00007680682,0.0001923208,0.0002319991,0.00007931343],"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":[0.000712991,0.000245449,0.00323251,0.0002091489,0.0001604304,0.000210238,0.0001668935,0.6397496,0.01732536,0.007325673,0.009670317,0.3209915],"study_design_scores_gemma":[0.00001726953,0.00003312476,0.0001597194,0.000005158573,0.000007495537,0.0000201191,0.00002138952,0.9936554,0.003926353,0.001622509,0.000524681,0.000006776317],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05819545,0.0004039249,0.9262565,0.0002318831,0.0001180486,0.0002566098,0.0004866037,0.01088367,0.00316728],"genre_scores_gemma":[0.4637862,0.0001519026,0.5298633,0.0002290058,0.00003630368,0.0003566257,0.001835522,0.0007831184,0.002958028],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01300392,"threshold_uncertainty_score":0.02585644,"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."}}